Master-slave game-based economic dispatching method for producers and diasters and shared energy storage

Abstract

The invention discloses a producer and consumer and shared energy storage economic dispatching method based on a master-slave game, and belongs to the technical field of power system economic dispatching operation. The scheduling method comprises the following steps: firstly, establishing a shared energy storage charging and discharging scheduling model; secondly, establishing a producer and consumer electric quantity balance scheduling model which is responsible for solving the problem of optimal scheduling of renewable energy consumption of producers and consumer and meeting of load requirements, and the optimization target is that the operation cost of the producers and consumer is minimum; and thirdly, establishing a producer and disappearing person and shared energy storage economic dispatching model based on the master-slave game. And 4, converting the double-layer optimization model into a single-layer optimization model by using a KKT condition, and linearizing the nonlinear constraint by using a Big-M method. And finally, establishing a shared energy storage and consumer internal transaction price updating model. According to the invention, an optimized operation strategy method is established for the shared energy storage and the producer, the net load between the producer and the consumer can be communicated, the producer and the consumer can be promoted to improve the utilization rate of renewable energy power generation, and the economic dispatching of the shared energy storage is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system dispatching and operation, and relates to a dispatching method for energy storage equipment under a market environment, and in particular to an economic dispatching method for prosumers and shared energy storage based on a master-slave game. Background Art

[0002] As greenhouse gas emissions increase and global temperatures continue to rise, building a new power system, promoting energy structure transformation, and reducing greenhouse gas emissions have become important issues in the energy and power industry, and renewable energy has been vigorously developed. Therefore, renewable energy technologies such as distributed energy such as rooftop solar and wind turbines have become more common. This has also led to the evolution of some traditional users into prosumers that integrate energy production and consumption, which can be both buyers and sellers in the electricity market. The emergence of prosumers gives new power systems more possibilities, but their small power generation scale and fluctuating net load characteristics will also bring new challenges to the safe and stable operation of the power system.

[0003] Energy storage has great advantages in tracking new energy power generation plans and participating in the peak-valley arbitrage market due to its high control accuracy and fast response speed. However, energy storage currently has disadvantages such as low utilization, single energy storage pricing and revenue model, and difficulty in recovering costs. The introduction of the sharing model provides a new solution for the operation of energy storage power stations, which can solve the shortcomings of high cost of energy storage power stations and long cost recovery period. The interaction between shared energy storage and renewable energy can improve the absorption rate of renewable energy and the utilization rate of energy storage resources.

[0004] However, due to the lag of market rules, the dispatching and operation of energy storage power stations in the sharing mode still have problems such as insufficient utilization, poor cost relief mechanism, unclear business model, and low marketization, resulting in technical bottlenecks in shared energy storage from dispatching mode to strategy selection. To solve this problem, existing research mainly provides solutions from the perspectives of setting the shared price of energy storage, energy storage charging and discharging decision-making, and multi-market player game. However, the existing solutions are not comprehensive enough in exploring the power market that shared energy storage can participate in, which will affect the dispatching decision and dispatching plan formulation of shared energy storage, and thus affect the power balance of the power system.

[0005] Therefore, based on the existing electricity market background and shared operation mode, it is urgent to propose a charging and discharging scheduling strategy suitable for shared energy storage power stations to ensure the profit level of shared energy storage while supporting the operation of the power system. Summary of the invention

[0006] In view of the technical problems faced by the optimal scheduling of shared energy storage under the market background, a method for the operation scheduling of prosumers and shared energy storage under a sharing mechanism is provided. When modeling the scheduling of shared energy storage, considering the market-based operation revenue of energy storage power stations facing multiple renewable energy prosumers under the market rules based on the sharing mechanism, a charging and discharging scheduling decision strategy for shared energy storage is formulated. On the premise of ensuring the load demand of prosumers and the consumption of renewable energy, the economic scheduling of shared energy storage is realized.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows:

[0008] An economic scheduling method for prosumers and shared energy storage based on a master-slave game. First, a charging and discharging scheduling model for shared energy storage is established. The charging and discharging scheduling model for shared energy storage is responsible for solving the charging and discharging decision problem of the shared energy storage power station in the electricity market environment, and the optimization objective is to maximize the daily operation revenue of the shared energy storage participating in the electricity market. Second, a power balance scheduling model for prosumers is established. The power balance scheduling model for prosumers is responsible for solving the optimal scheduling problem of prosumer renewable energy consumption and meeting load demand, and the optimization objective is to minimize the operation cost of prosumers. Third, an economic scheduling model for prosumers and shared energy storage based on a master-slave game is established. Fourth, the double-layer optimization model is transformed into a single-layer optimization model using the KKT conditions, and the Big-M method is used to linearize the non-linear constraints. Finally, an internal trading price update model for shared energy storage and prosumers is established. It includes the following steps:

[0009] The first step is to establish a charging and discharging scheduling model for shared energy storage;

[0010] The charging and discharging scheduling model for shared energy storage is used to solve the charging and discharging decision problem of the shared energy storage power station in the electricity market environment, and the optimization objective is to maximize the daily operation revenue of the shared energy storage participating in the electricity market. It includes the trading electricity volume with prosumers in the day-ahead energy market, the capacity of the capacity lease transaction with prosumers, and the electricity volume of the power purchase and sale transaction with the power grid.

[0011] The objective function of the charging and discharging scheduling model for shared energy storage is shown in Equation (1):

[0012] maxF = F ses +F cap -F grid -F equipment (1)

[0013] In the formula, F is the daily operation revenue of the shared energy storage; F ses is the energy lease revenue of the shared energy storage, as shown in Equation (2); F cap is the reserve selling revenue of the shared energy storage, as shown in Equation (3); F grid is the revenue of the shared energy storage in the transaction with the power grid, as shown in Equation (4); F equipmentFor the construction and operation and maintenance costs of shared energy storage, as shown in Equation (5).

[0014]

[0015] In the formula, T is the scheduling period; I is the set of prosumers; p sell,t and p buy,t are the selling price and purchasing price of the shared energy storage; P ses,sell,i,t and P ses,buy,i,t are the selling electricity volume and purchasing electricity volume of the shared energy storage in the energy field; t represents the moment t; i represents the i-th prosumer; p bu,t and p bd,t are the upper and lower reserve selling prices of the shared energy storage; P cap,ui,t and P cap,di,t are the upper and lower reserve selling volumes of the shared energy storage; ΔP cap,ui,t and ΔP cap,di,t are the actual usage amounts of the reserve sold by the shared energy storage to the prosumers; p gbuy,t and p gsell,t are the prices of purchasing electricity from and selling electricity to the superior power grid by the shared energy storage respectively; P gb,t and P gs,t are the power of purchasing electricity from and selling electricity to the superior power grid by the shared energy storage respectively; π c is the capacity cost of the shared energy storage; π a is the energy cost of the shared energy storage; π b is the operation and maintenance cost of the shared energy storage; time is the theoretical life cycle of the energy storage; S cap is the rated capacity of the energy storage; P socmax is the maximum charge and discharge power of the shared energy storage.

[0016] The constraint conditions of the charge and discharge scheduling model of the shared energy storage are as follows:

[0017] (1) The balance constraint between the charge and discharge and trading power of the shared energy storage, as shown in Equation (6).

[0018]

[0019] In the formula, P ch,t and P dis,t are the charge and discharge powers of the shared energy storage respectively; the other variables are the same as those described above.

[0020] (2) The charge and discharge power constraint of the shared energy storage, as shown in Equation (7).

[0021]

[0022] In the formula, U ch,t is the charge and discharge state of the shared energy storage. When the value is 1, it represents the charging state; when the value is 0, it represents the discharging state; the other variables are the same as those described above.

[0023] (3) The trading electricity quantity constraint between the shared energy storage and the superior power grid is shown in Equation (8).

[0024]

[0025] In the formula, U gs,t represents the trading status between the shared energy storage and the superior power grid. When the value is 1, it means the shared energy storage sells to the superior power grid; when the value is 0, it means the shared energy storage purchases from the superior power grid; M is a large number; the other variables are the same as those described above.

[0026] (4) The storage capacity constraint of the shared energy storage is shown in Equation (9).

[0027]

[0028] In the formula, S max and S min are the maximum and minimum stored electricity quantities of the shared energy storage respectively; S t and S t+1 correspond to the stored electricity quantities at time t and time t + 1 respectively; Δt is the scheduling period interval; the other variables are the same as those described above.

[0029] (5) The capacity balance constraint within the scheduling period of the shared energy storage is shown in Equation (10).

[0030] S0 = S T (10)

[0031] In the formula, S0 is the initial stored electricity quantity of the shared energy storage; S T is the stored electricity quantity at the end of the scheduling period of the shared energy storage.

[0032] Second, establish the prosumer electricity balance scheduling model; the prosumer electricity balance scheduling model is responsible for solving the optimal scheduling problem of prosumer renewable energy consumption and meeting load demand, and the optimization goal is to minimize the prosumer operation cost. Its operation cost includes the transaction cost with the shared energy storage, the transaction cost with the power grid, and the deviation penalty cost given by the power grid when the intraday load cannot be met.

[0033] The objective function of the prosumer electricity balance scheduling model is:

[0034] minC i = C ses,i + C cap.i + C grid.i + C punish,i (11)

[0035] In the formula, C i is the daily operation cost of prosumer i, as shown in Equation (12); C ses,i is the energy leasing cost of prosumer i, as shown in Equation (13); Ccap.i is the backup purchase cost of prosumer i, as shown in Equation (14); C grid.i is the cost of prosumer i's transaction with the power grid, as shown in Equation (15); C punish,i is the deviation penalty given by the power grid to prosumer i when the intraday load cannot be met, as shown in Equation (16).

[0036]

[0037] In the formula, P pgb,i,t and P pgs,i,t are the power sold by prosumer i to the power grid and purchased from the power grid at time t, respectively; π + and π - are the upper and lower deviation penalty factors, respectively; ΔP pgb,i,t and ΔP pgs,i,t are the upper and lower deviation amounts, respectively, as shown in Equation (17), and can be further expressed as Equation (18) and Equation (19); the remaining variables are the same as those described above.

[0038]

[0039] In the formula, P Li,t is the predicted load of prosumer i; P i,t is the generation dispatch power of the distributed power source in prosumer i, and should satisfy Equation (20); U 1,i,t , U 2,i,t , U 3,i,t , U 4,i,t are auxiliary variables, taking values of 0 or 1; the remaining variables are the same as those described above.

[0040]

[0041] In the formula, is the predicted power generation of the distributed power source in prosumer i; is the power generation prediction deviation factor of the distributed power source in prosumer i.

[0042] The constraint conditions of the prosumer power balance dispatch model are as follows:

[0043] (1) Prosumer transaction balance constraint, as shown in Equation (21).

[0044] P Li,t -P i,t =P ses,sell,i,t -P ses,buy,i,t +P pgb,i,t -P pgs , i, t (21)

[0045] In the formula, the variables are the same as those described above.

[0046] (2) The prosumer's reserve purchase quantity constraint, as shown in equations (22), (23), (24), and (25).

[0047]

[0048] Wherein, the variables are the same as those described above.

[0049] (3) The prosumer's reserve actual usage quantity constraint, as shown in equations (26) and (27).

[0050]

[0051] Wherein, U cap,i,t represents the usage status of the reserve. When the value is 1, it means the upper reserve is used; when the value is 0, it means the lower reserve is used; the other variables are the same as those described above.

[0052] (4) The prosumer's transaction volume constraint with the superior power grid, as shown in equation (28).

[0053]

[0054] Wherein, U pg,i,t represents the transaction status with the superior power grid. When the value is 1, it means selling to the power grid; when the value is 0, it means purchasing from the power grid; the other variables are the same as those described above.

[0055] (5) The prosumer's transaction volume constraint with the shared energy storage, as shown in equations (29) and (30).

[0056]

[0057] Wherein, U ses,i,t represents the transaction status between the prosumer and the shared energy storage. When the value is 1, it means the prosumer purchases power from the shared energy storage; when the value is 0, it means the prosumer sells power to the shared energy storage; the other variables are the same as those described above.

[0058] Step 3: Establish a prosumer and shared energy storage economic dispatch model based on the master-slave game;

[0059] The shared energy storage charge-discharge dispatch model and the prosumer power balance dispatch model form a master-slave game, forming a two-layer dispatch model: the outer master game is the shared energy storage charge-discharge dispatch model, and the inner slave game is the prosumer power balance dispatch model.

[0060] The objective function of the outer layer model is as shown in equation (1), and the constraint conditions include equations (6), (7), (8), (9), and (10). The variables to be optimized in the outer layer include P gb,t 、P gs,t; The objective function of the inner-layer model is as shown in Equation (11), and the constraint conditions include Equation (18), Equation (19), Equation (20), Equation (21), Equation (22), Equation (23), Equation (24), Equation (25), Equation (26), Equation (27), Equation (28), Equation (29), Equation (30). The variables to be optimized in the inner layer include P pgb,i,t , P pgs,i,t , P cap,ui,t , P cap,di,t , P ses,sell,i,t , P ses,buy,i,t , ΔP cap,ui,t , ΔP cap.di,t .

[0061] Step 4: Use the KKT conditions to transform the bilevel optimization model into a single-level optimization model;

[0062] For the shared energy storage charge and discharge scheduling model and prosumer power balance scheduling model based on the master-slave game, there is a coupling relationship between the master game and the slave game models, and there are integer variables, continuous variables, and non-linear constraints, making it difficult to directly optimize and solve. In the present invention, a Lagrangian function is constructed for the inner-layer slave game. Based on the KKT complementary slackness conditions, the inner-layer slave game is transformed into the constraint conditions of the master game model, so that the master-slave game is transformed into a single-level non-linear model. And the Big-M method is used to linearize the non-linear constraints.

[0063] Step 5: Establish an update model for the internal trading price between the shared energy storage and the prosumer;

[0064] Solve the single-level linear model of the mixed-integer linear programming problem after equivalent transformation. Based on the storage and discharge capacity of the shared energy storage and the trading demand of the prosumer, calculate and update the internal trading price between the shared energy storage and the prosumer, as shown in Equation (87).

[0065]

[0066] In the formula, p buy,t , p sell,t are the internal trading prices when the shared energy storage purchases and sells electricity to the prosumer, respectively; σ a,t , σ b,t are the price factors when the shared energy storage sells and purchases electricity to the prosumer, respectively, as shown in Equation (88); the other variables are the same as those described above.

[0067]

[0068] In the formula, S a,t is the state of the stored electricity of the shared energy storage after trading with the power grid, as shown in Equation (89); the other variables are the same as those described above.

[0069] S a,t+1 = S a,t+P gb,t -P gs,t (89)

[0070] Wherein, the variables are the same as those described above.

[0071] After each optimization, the shared energy storage trading price is iteratively updated until the results of two price iterations remain unchanged, which means convergence. The charging and discharging scheduling result of the shared energy storage obtained under the convergence condition is the economic scheduling optimization solution sought.

[0072] The effects and benefits of the present invention are as follows:

[0073] Aiming at the technical problems faced by the optimal scheduling of shared energy storage in the market background, the present invention provides an operation scheduling method for prosumers and shared energy storage under a sharing mechanism. When modeling the shared energy storage scheduling, considering the market operation income of the energy storage power station facing multiple renewable energy prosumers under the market rules based on the sharing mechanism, a decision-making strategy for the charging and discharging scheduling of the shared energy storage is formulated. The proposed master-slave game model can communicate the net load among prosumers, encourage prosumers to improve the utilization rate of renewable energy power generation, and reduce the volatility brought to the power grid by the individual grid connection of prosumers. On the premise of ensuring the load demand of prosumers and the consumption of renewable energy, the economic scheduling of shared energy storage is realized, and the income level of shared energy storage is guaranteed. Brief Description of the Drawings

[0074] Figure 1 It is a flowchart of an economic scheduling method for prosumers and shared energy storage based on master-slave game.

[0075] Figure 2 It is a running framework diagram of shared energy storage - prosumer.

[0076] Figure 3 It is the net load diagram of the prosumer.

[0077] Figure 4 It is the convergence result diagram.

[0078] Figure 5 It is the prosumer's reserve purchase diagram.

[0079] Figure 6 It is the power selling diagram of shared energy storage - prosumer.

[0080] Figure 7 It is the power purchase diagram of shared energy storage - prosumer.

[0081] Figure 8 It is the power purchase diagram of shared energy storage - power grid.

[0082] Figure 9 It is the power selling diagram of shared energy storage - power grid. Detailed Embodiments

[0083] The following takes the simulation based on the previous 24-hour period as an example. The example includes a shared energy storage, a PV prosumer, and a wind power prosumer. The net load of the prosumer is as Figure 3 , and the specific implementation process of the present invention will be described in detail in combination with the technical solution and the attached drawings. The specific implementation process is as follows: Table 1 shows the parameters of the shared energy storage.

[0084] First step, establish a charging and discharging scheduling model for the shared energy storage;

[0085] The charging and discharging scheduling model for the shared energy storage is used to solve the charging and discharging decision-making problem of the shared energy storage power station in the electricity market environment, and the optimization goal is to maximize the daily operating income of the shared energy storage participating in the electricity market. It includes the trading electricity volume with the prosumer in the day-ahead energy market, the capacity of the capacity leasing transaction with the prosumer, and the electricity volume of the power purchase and sale transaction with the power grid.

[0086] The objective function of the charging and discharging scheduling model of the shared energy storage is shown in Equation (1): In Equation (1), F is the daily operating income of the shared energy storage; F ses is the energy leasing income of the shared energy storage, as shown in Equation (2); F cap is the standby selling income of the shared energy storage, as shown in Equation (3); F grid is the income of the shared energy storage from trading with the power grid, as shown in Equation (4); F equipment is the construction and operation and maintenance cost of the shared energy storage, as shown in Equation (5). In Equation (5), T is the scheduling period; I is the set of prosumers; p sell,t , p buy,t are the selling price and purchase price of the shared energy storage; P ses,sell,i,t , P ses,buy,i,t are the selling electricity volume and purchase electricity volume of the shared energy storage in the energy field; t represents the moment t; i represents the i-th prosumer; p bu,t , p bd,t are the upper and lower standby selling prices of the shared energy storage; P cap,ui,t , P cap,di,t are the upper and lower standby selling volumes of the shared energy storage; ΔP cap,ui,t , ΔP cap,di,t are the actual usage amounts of the standby sold by the shared energy storage to the prosumer; p gbuy,t , p gsell,t are the prices of purchasing electricity from and selling electricity to the superior power grid by the shared energy storage respectively; P gb,t , P gs,t are the power of purchasing electricity from and selling electricity to the superior power grid by the shared energy storage respectively; π c is the capacity cost of the shared energy storage; π a is the energy cost of the shared energy storage; π b is the operation and maintenance cost of the shared energy storage; time is the theoretical life cycle of the energy storage; S cap is the rated capacity of the energy storage; P socmaxFor the maximum charge and discharge power of shared energy storage.

[0087] The constraint conditions of the shared energy storage charge and discharge scheduling model are as follows:

[0088] (1) The balance constraint between the shared energy storage charge and discharge and the trading power, as shown in Equation (6). In Equation (6), P ch,t , P dis,t are the charging and discharging powers of the shared energy storage respectively; the other variables are the same as those described above.

[0089] (2) The shared energy storage charge and discharge power constraint, as shown in Equation (7). In Equation (7), U ch,t is the charge and discharge state of the shared energy storage. When the value is 1, it represents the charging state; when the value is 0, it represents the discharging state; the other variables are the same as those described above.

[0090] (3) The constraint on the electricity trading volume between the shared energy storage and the superior power grid, as shown in Equation (8). In Equation (8), U gs,t represents the trading state between the shared energy storage and the superior power grid. When the value is 1, it means the shared energy storage sells to the superior power grid; when the value is 0, it means the shared energy storage purchases from the superior power grid; M is a large number; the other variables are the same as those described above.

[0091] (4) The shared energy storage storage capacity constraint, as shown in Equation (9). In Equation (9), S max , S min are the maximum and minimum stored electricity of the shared energy storage respectively; S t , S t+1 correspond to the stored electricity at time t and time t + 1 respectively; Δt is the scheduling time interval; the other variables are the same as those described above.

[0092] (4) The capacity balance constraint within the scheduling period of the shared energy storage, as shown in Equation (10). In Equation (10), S0 is the initial stored electricity of the shared energy storage; S T is the stored electricity of the shared energy storage at the end of the scheduling period.

[0093] Second step, establish the prosumer electricity balance scheduling model; the prosumer electricity balance scheduling model is responsible for solving the optimal scheduling problem of prosumer renewable energy consumption and meeting load demand, and the optimization goal is to minimize the prosumer operation cost. Its operation cost includes the transaction cost with the shared energy storage, the transaction cost with the power grid, and the deviation penalty cost given by the power grid when the daily load cannot be met.

[0094] The objective function of the prosumer electricity balance scheduling model is: In Equation (11), C i is the daily operation cost of prosumer i, as shown in Equation (12); C ses,i is the energy leasing cost of prosumer i, as shown in Equation (13); C cap.iis the backup purchase cost of prosumer i, as shown in Equation (14); C grid.i is the cost of prosumer i's transaction with the power grid, as shown in Equation (15); C punish,i is the deviation penalty given by the power grid to prosumer i when the intraday load cannot be met, as shown in Equation (16). In Equation (16), P pgb,i,t and P pgs,i,t are the power sold by and purchased by prosumer i from the power grid at time t, respectively; π + and π - are the upper and lower deviation penalty factors, respectively; ΔP pgb,i,t and ΔP pgs,i,t are the upper and lower deviation amounts, respectively, as shown in Equation (17), and can be further expressed as Equations (18) and (19); the other variables are the same as those described above. In Equation (19), P Li,t is the predicted load of prosumer i; P i,t is the power generation scheduling power of the distributed power source in prosumer i, which should satisfy Equation (20); U 1,i,t and U 2,i,t and U 3,i,t and U 4,i,t are auxiliary variables, taking values of 0 or 1; the other variables are the same as those described above. In Equation (20), is the predicted power generation of the distributed power source in prosumer i; is the power generation prediction deviation factor of the distributed power source in prosumer i.

[0095] The constraint conditions of the prosumer's power balance scheduling model are as follows:

[0096] (1) Prosumer transaction balance constraint, as shown in Equation (21). In Equation (21), the variables are the same as those described above.

[0097] (2) Prosumer backup purchase quantity constraint, as shown in Equations (22), (23), (24), and (25). In Equations (22), (23), (24), and (25), the variables are the same as those described above.

[0098] (3) Prosumer backup actual usage quantity constraint, as shown in Equations (26) and (27). In Equations (26) and (27), U cap,i,t represents the usage status of the backup. When the value is 1, it means the upper backup is used; when the value is 0, it means the lower backup is used; the other variables are the same as those described above.

[0099] (4) Prosumer's transaction volume constraint with the superior power grid, as shown in Equation (28). In Equation (20), U pg,i,t represents the transaction status with the superior power grid. When the value is 1, it means selling to the power grid; when the value is 0, it means purchasing from the power grid; the other variables are the same as those described above.

[0100] (5) The trading volume constraint between prosumers and shared energy storage is shown in Equations (29) and (30). In Equations (29) and (30), U ses,i,t represents the trading status between prosumers and shared energy storage. When the value is 1, it means the prosumer purchases power from the shared energy storage; when the value is 0, it means the prosumer sells power to the shared energy storage; the other variables are the same as those described above.

[0101] It can be seen from Figure 5 that since the purchase of prosumer reserve is generated by the volatility of renewable energy generation, the prosumer only purchases reserve when there is configured renewable energy generation. The purchase of reserve is also limited by the SOC of the shared energy storage, so the reserve purchase volume at individual time points is 0. For prosumers, both upward and downward deviations will be penalized. However, the prosumer's perception of being penalized for the generation of downward deviation will be stronger than the upward deviation generated when the load is not met. Therefore, the prosumer will make a choice to purchase a large amount of upward reserve and hardly purchase downward reserve.

[0102] Step 3: Establish an economic dispatch model for prosumers and shared energy storage based on the master-slave game;

[0103] The charging and discharging dispatch model of the shared energy storage and the prosumer's power balance dispatch model form a master-slave game, forming a two-layer dispatch model: the outer master game is the charging and discharging dispatch model of the shared energy storage, and the inner slave game is the prosumer's power balance dispatch model. The operation framework diagram of the prosumer and shared energy storage economic dispatch model is as shown in Figure 2 shown.

[0104] The objective function of the outer layer model is shown in Equation (1), and the constraint conditions include Equations (6), (7), (8), (9), (10). The variables to be optimized in the outer layer include P gb,t , P gs,t ; the objective function of the inner layer model is shown in Equation (11), and the constraint conditions include Equations (18), (19), (20), (21), (22), (23), (24), (25), (26), (27), (28), (29), (30). The variables to be optimized in the inner layer include P pgb,i,t , P pgs,i,t , P cap,ui,t , P cap,di,t , P ses,sell,i,t , P ses,buy,i,t , ΔP cap,ui,t , ΔP cap.di,t .

[0105] Step 4: Use the KKT conditions to transform the two-layer optimization model into a single-layer optimization model;

[0106] For the above-mentioned energy storage charge and discharge scheduling model and prosumer power balance scheduling model based on master-slave game, there is a coupling relationship between the master game and the slave game models, and there are integer variables, continuous variables and non-linear constraints, which are difficult to directly optimize and solve. In the present invention, the Lagrangian function is constructed for the inner-layer slave game. Based on the KKT complementary slackness conditions, the inner-layer slave game is converted into the constraint conditions of the master game model, so that the master-slave game is converted into a single-layer non-linear model. And the Big-M method is used to linearize the non-linear constraints. The solving process is as follows: First, the Lagrangian function of the prosumer is constructed through the Lagrangian factors of the constraint equations, and the equivalent constraints are obtained by taking the gradient of the Lagrangian function. Finally, the Big-M method is used to equivalently transform the original non-linear constraints into mixed integer linear constraints through 0-1 variables. The specific solving process is shown in 4.1-4.6:

[0107] Step 4.1, take the Lagrangian factors of the equality constraint equations (21) and (30) as μ 1,i,t , μ 2,i,t ; the Lagrangian factors of the inequality constraint equation (18) are λ 1,i,t , λ 2,i,t , λ 3,i,t , λ 4,i,t , λ 5,i,t ; the Lagrangian factors of the equation (19) are λ 6,i,t , λ 7,i,t , λ 8,i,t , λ 9,i,t , λ 10,i,t ; the Lagrangian factors of the equation (20) are λ 11,i,t , λ 12,i,t ; the Lagrangian factors of the equation (22) are λ 13,i,t , λ 14,i,t ; the Lagrangian factors of the equation (23) are λ 15,i,t , λ 16,i,t ; the Lagrangian factors of the equation (24) are λ 17,i,t , λ 18,i,t ; the Lagrangian factors of the equation (25) are λ 19,i,t , λ 20,i,t ; the Lagrangian factors of the equation (26) are λ 21,i,t , λ 22,i,t ; the Lagrangian factors of the equation (27) are λ 23,i,t , λ 24,i,t , λ 25,i,t , λ 26,i,t ; the Lagrangian factors of the equation (28) are λ 27,i,t , λ 28,i,t , λ 29,i,t , λ 30,i,t ; the Lagrangian factors of the equation (29) are λ 31,i,t , λ 32,i,t , λ 33,i,t, λ 34,i,t .

[0108] Step 4.2, construct the Lagrangian function of the prosumer

[0109]

[0110] In the formula, the variables are the same as those described above.

[0111] Step 4.3, take the gradient of the Lagrangian function to obtain the equivalent constraints, such as Equations (32) - (52)

[0112]

[0113] μ 1,i,t + λ 2,i,t - λ 4,i,t - λ 7,i,t + λ 9,i,t - λ 11,i,t + λ 12,i,t + λ 13,i,t + λ 14.i.t = 0 (37)

[0114] - Mλ 24,i,t + Mλ 26,i,t = 0 (38)

[0115] - Mλ 28,i,t + Mλ 30,i,t = 0 (39)

[0116] - P socmax λ 32,i,t + P socmax λ 34,i,t = 0 (40)

[0117]

[0118]

[0119] In the formula, "0 ≤ a ⊥ b ≥ 0" means "a ≥ 0, b ≥ 0 and ab = 0"; each variable in the formula is the same as those described above.

[0120] Step 4.4, adopt the Big - M method, and through 0 - 1 variables, equivalently transform the original non - linear constraints into mixed - integer linear constraints. Equations (42) - (52) can be equivalently transformed into Equations (53) - (86).

[0121]

[0122]

[0123]

[0124]

[0125] In the formula, M2 is the large number introduced by the Big-M method; ν 1.i.t , ν 2.i.t , ν 3.i.t , ν 4.i.t , ν 5.i.t , ν 6.i.t , ν 7.i.t , ν 8.i.t , ν 9.i.t , ν 10.i.t , ν 11.i.t , ν 12.i.t , ν 13.i.t , ν 14.i.t , ν 15.i.t , ν 16.i.t , ν 17.i.t , ν 18.i.t , ν 19.i.t , ν 20.i.t , ν 21.i.t , ν 22.i.t , ν 23.i.t , ν 24.i.t , ν 25.i.t , ν 26.i.t , ν 27.i.t , ν 28.i.t , ν 29.i.t , ν 30.i.t , ν 31.i.t , ν 32.i.t , ν 33.i.t , ν 34.i.t are the 0-1 variables introduced by the Big-M method; the remaining variables are the same as those described above.

[0126] Step 4.5, the constraint conditions after the equivalence of the inner-layer model include Equation (21), Equation (30), Equations (32) to (41), and Equations (53) to (86).

[0127] Step 4.6, for the single-layer nonlinear model of the mixed-integer linear programming problem after equivalent conversion, the objective function is as shown in Equation (1), and the constraint conditions include Equations (6) to (10), Equation (21), Equation (30), Equations (32) to (41), and Equations (53) to (86), which are solved by calling the commercial solver CPLEX and the YALMIP toolbox in MATLAB.

[0128] From Figure 6 and Figure 7It can be seen that shared energy storage tends to purchase electricity from prosumers at a lower price when there is a choice, and sell most of the electricity sold at the price peak to prosumers. When the net load curves of two prosumers are complementary, shared energy storage can gain benefits by acting as a middleman to communicate the net loads of the two prosumers.

[0129] From Figure 8 and Figure 9 , it can be concluded that shared energy storage sells at a high price and buys at a low price in the transaction with the power grid, and can utilize the characteristics of the time-of-use electricity price of the power grid for arbitrage.

[0130] Step 5: Establish an internal trading price update model for shared energy storage and prosumers;

[0131] Solve the single-layer linear model of the mixed-integer linear programming problem after equivalent conversion. Based on the storage and discharge capacity of shared energy storage and the trading demands of prosumers, calculate and update the internal trading price between shared energy storage and prosumers, as shown in Equation (87).

[0132]

[0133] In the formula, p buy,t , p sell,t are the internal trading prices when shared energy storage purchases and sells electricity to prosumers respectively; σ a,t , σ b,t are the price factors when shared energy storage sells and purchases electricity to prosumers respectively, as shown in Equation (88); the other variables are the same as those described above.

[0134]

[0135] In the formula, S a,t is the state of the stored electricity amount after the transaction between shared energy storage and the power grid, as shown in Equation (89); the other variables are the same as those described above.

[0136] S a,t+1 =S a,t +P gb,t -P gs,t (89)

[0137] In the formula, the variables are the same as those described above.

[0138] Each time after optimization, iterate and update the trading price of shared energy storage until the results of two price iterations remain unchanged, which means convergence. The charging and discharging scheduling result of shared energy storage obtained under the convergence condition is the economic scheduling optimization solution sought. As Figure 4As shown, it is the variation law of the difference in the daily income of shared energy storage in the iteration, with the daily income of shared energy storage only trading with the power grid as the starting point. From the relatively large initial difference, it can be concluded that shared energy storage can obtain greater benefits by trading with both the power grid and prosumers at the same time. Since the optimization result of the previous iteration will cause changes in the power purchase and sale prices of the subsequent iteration, the difference will fluctuate during the iteration process. As the number of iterations continues to increase, after the 27th optimization, the program converges to obtain the final optimization result.

[0139] Table 1 Configuration parameters of shared energy storage

[0140]

[0141] The above-described embodiments only represent the implementation modes of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that for those skilled in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. A method for economic dispatching of prosumers and shared energy storage based on master-slave game, characterized in that The prosumer and shared energy storage economic dispatch method includes the following steps: In the first step, a shared energy storage charging and discharging dispatch model is established. The shared energy storage charging and discharging dispatch model is used to solve the charging and discharging decision-making problem of the shared energy storage power station in the electricity market environment. The optimization objective is to maximize the daily operating income of the shared energy storage participating in the electricity market, including the transaction electricity volume with the prosumer in the day-ahead energy market, the capacity of the capacity lease transaction with the prosumer, and the electricity volume of the power purchase and sale transaction with the power grid; In the second step, a prosumer electricity balance dispatch model is established; the prosumer electricity balance dispatch model is responsible for solving the optimal dispatch problem of prosumer renewable energy consumption and meeting load demand. The optimization objective is to minimize the operating cost of the prosumer; its operating cost includes the transaction cost with the shared energy storage, the transaction cost with the power grid, and the deviation penalty cost given by the power grid when the intraday load cannot be met; In the third step, a prosumer and shared energy storage economic dispatch model based on master-slave game is established; In the fourth step, the double-layer optimization model is transformed into a single-layer optimization model by using the KKT conditions, and the Big-M method is used to linearize the non-linear constraints; In the fifth step, a shared energy storage and prosumer internal trading price update model is established, and the obtained shared energy storage charging and discharging dispatch result after solving it is the economic dispatch optimization solution sought.

2. The economic dispatch method for prosumers and shared energy storage based on master-slave game according to claim 1, wherein The first step described above includes the following steps: The objective function of the shared energy storage charging and discharging dispatch model is shown in Equation (1): maxF = F ses +F cap -F grid -F equipment (1) In the formula, F is the daily operating income of the shared energy storage; F ses is the energy leasing income of the shared energy storage, as shown in Equation (2); F cap is the standby selling income of the shared energy storage, as shown in Equation (3); F grid is the income from the transaction between the shared energy storage and the power grid, as shown in Equation (4); F equipment is the construction and operation and maintenance cost of the shared energy storage, as shown in Equation (5); where T is the scheduling period; I is the set of prosumers; p sell,t and p buy,t are the selling price and the purchasing price of the shared energy storage; P ses,sell,i,t and P ses,buy,i,t are the selling electricity quantity and the purchasing electricity quantity of the shared energy storage in the energy field; t represents the moment t; i represents the i-th prosumer; p bu,t and p bd,t are the upper and lower reserve selling prices of the shared energy storage; P cap,ui,t and P cap,di,t are the upper and lower reserve selling quantities of the shared energy storage; ΔP cap,ui,t and ΔP cap,di,t are the actual usage amounts of the reserve sold by the shared energy storage to the prosumers; p gbuy,t and p gsell,t are the prices of the shared energy storage for purchasing electricity from and selling electricity to the superior power grid respectively; P gb,t and P gs,t are the power of the shared energy storage for purchasing electricity from and selling electricity to the superior power grid respectively; π c is the capacity cost of the shared energy storage; π a is the energy cost of the shared energy storage; π b is the operation and maintenance cost of the shared energy storage; time is the theoretical life cycle of the energy storage; S cap is the rated capacity of the energy storage; P socmax is the maximum charge-discharge power of the shared energy storage; The constraint conditions of the shared energy storage charging and discharging dispatch model include the balance constraint between the shared energy storage charging and discharging and the transaction power, the shared energy storage charging and discharging power constraint, the shared energy storage and the upper-level power grid transaction electricity volume constraint, the shared energy storage storage capacity constraint, and the capacity balance constraint within the shared energy storage dispatch period.

3. The economic dispatch method for prosumers and shared energy storage based on master-slave game according to claim 2, characterized in that, The constraint conditions of the shared energy storage charging and discharging dispatch model described above are specifically: (1) The balance constraint between the shared energy storage charging and discharging and the transaction power, as shown in Equation (6); Wherein, P ch,t and P dis,t are the charging and discharging powers of the shared energy storage respectively; (2) The shared energy storage charging and discharging power constraint, as shown in Equation (7); Where U ch,t is the charge and discharge state of the shared energy storage. When the value is 1, it represents the charging state; when the value is 0, it represents the discharging state. (3) The shared energy storage and the upper-level power grid transaction electricity volume constraint, as shown in Equation (8); Where U gs,t represents the trading status of the shared energy storage with the superior power grid. When the value is 1, it means the shared energy storage sells to the superior power grid; when the value is 0, it means the shared energy storage purchases from the superior power grid. M is a large number; (4) The shared energy storage storage capacity constraint, as shown in Equation (9); where S max , S min are the maximum and minimum stored energies of the shared energy storage, respectively; S t , S t+1 correspond to the stored energies at time t and time t + 1, respectively; Δt is the scheduling time interval; (5) The capacity balance constraint within the shared energy storage dispatch period, as shown in Equation (10); S0 = S T (10) Wherein, S0 is the initial stored power of the shared energy storage; S T is the stored power at the end of the scheduling period of the shared energy storage.

4. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 1, characterized in that The second step described above includes the following steps: The objective function of the prosumer electricity balance dispatch model is: minC i = C ses,i + C cap.i + C grid.i + C punish,i (11) where, C i is the daily operating cost of prosumer i, as shown in Equation (12); C ses,i is the energy leasing cost of prosumer i, as shown in Equation (13); C cap.i is the reserve purchase cost of prosumer i, as shown in Equation (14); C grid.i is the cost of prosumer i's transaction with the power grid, as shown in Equation (15); C punish,i is the deviation penalty given by the power grid to prosumer i when the intraday load cannot be met, as shown in Equation (16). Where, P pgb,i,t and P pgs,i,t are the power sold by prosumer i to the power grid and purchased by prosumer i from the power grid during period t, respectively; π + and π - are the upper and lower deviation penalty factors, respectively; ΔP pgb,i,t and ΔP pgs,i,t are the upper and lower deviation amounts, respectively; The constraint conditions of the prosumer electricity balance dispatch model include the prosumer transaction balance constraint, the prosumer reserve purchase volume constraint, the prosumer reserve actual usage constraint, the prosumer and the upper-level power grid transaction volume constraint, and the prosumer and the shared energy storage transaction volume constraint.

5. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 4, characterized in that The above and below deviation amounts ΔP pgb,i,t and ΔP pgs,i,t As shown in Equation (17), it is further expressed as Equation (18) and Equation (19); where, P Li,t is the predicted load of prosumer i; P i,t is the power generation scheduling power of distributed power sources in prosumer i, and should satisfy Equation (20); U 1,i,t , U 2,i,t , U 3,i,t , U 4,i,t are auxiliary variables, and their values are 0 or 1; Wherein, is the predicted power generation of the distributed power source in prosumer i; is the power generation prediction deviation factor of the distributed power source in prosumer i.

6. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 4, characterized in that The constraint conditions of the prosumer electricity balance dispatch model are specifically: (1) The prosumer transaction balance constraint, as shown in Equation (21); P Li,t -P i,t =P ses,sell,i,t -P ses,buy,i,t +P pgb,i,t -P pgs,i,t (21) (2) The prosumer reserve purchase volume constraint, as shown in Equations (22), (23), (24), and (25); (3) The prosumer reserve actual usage constraint, as shown in Equations (26) and (27); Wherein, U cap,i,t represents the standby usage status. When the value is 1, it means the upper standby is in use; when the value is 0, it means the lower standby is in use; (4) The prosumer and the upper-level power grid transaction volume constraint, as shown in Equation (28); where U pg,i,t represents the trading status with the superior power grid. When the value is 1, it means selling to the power grid; when the value is 0, it means purchasing from the power grid. (5) The prosumer and the shared energy storage transaction volume constraint, as shown in Equations (29) and (30); Where, U ses,i,t represents the trading status between prosumers and shared energy storage. When the value is 1, it means that the prosumer purchases power from the shared energy storage; when the value is 0, it means that the prosumer sells power to the shared energy storage.

7. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claims 2, 3, 5, and 6, characterized in that The third step described above includes the following steps: The shared energy storage charge-discharge scheduling model and the prosumer power balance scheduling model form a master-slave game, resulting in a two-layer scheduling model: the outer master game is the shared energy storage charge-discharge scheduling model, and the inner slave game is the prosumer power balance scheduling model; The objective function of the outer layer model is shown in Equation (1), and the constraint conditions include Equation (6), Equation (7), Equation (8), Equation (9), and Equation (10). The variables to be optimized in the outer layer include P gb,t , P gs,t ; The objective function of the inner layer model is shown in Equation (11), and the constraint conditions include Equation (18), Equation (19), Equation (20), Equation (21), Equation (22), Equation (23), Equation (24), Equation (25), Equation (26), Equation (27), Equation (28), Equation (29), and Equation (30). The variables to be optimized in the inner layer include P pgb,i,t , P pgs,i,t , P cap,ui,t , P cap,di,t , P ses,sell,i,t , P ses,buy,i,t , ΔP cap,ui,t , ΔP cap.di,t .

8. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 7, characterized in that The fourth step described above includes the following steps: Solve the shared energy storage charge-discharge scheduling model and the prosumer power balance scheduling model based on the master-slave game; construct the Lagrangian function for the inner slave game, and based on the KKT complementary slackness conditions, convert the inner slave game into the constraint conditions of the master game model, so that the master-slave game is converted into a single-layer nonlinear model; and use the Big-M method to linearize the nonlinear constraints.

9. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 8, characterized in that The fifth step described above includes the following steps: Solve the single-layer linear model of the equivalent-converted mixed-integer linear programming problem, and calculate and update the internal trading prices between the shared energy storage and the prosumers based on the storage and discharge capacity of the shared energy storage and the trading demands of the prosumers, as shown in Equation (87); Where p buy,t and p sell,t are the internal trading prices when the shared energy storage purchases and sells electricity to prosumers respectively; σ a,t and σ b,t are the price factors when the shared energy storage sells and purchases electricity to prosumers respectively; Iteratively update the shared energy storage trading price after each optimization until the results of two price iterations remain unchanged, which means convergence. The obtained shared energy storage charge-discharge scheduling result under the convergence condition is the economic scheduling optimization solution sought.

10. A prosumer and shared energy storage economic dispatch method based on master-slave game according to claim 9, characterized in that, The described σ a,t , σ b,t As shown in Equation (88): where S a,t is the state of the stored electricity after the shared energy storage trades with the power grid, as shown in Equation (89); S a,t+1 = S a,t + P gb,t - P gs,t (89).

Images