Multi-virtual power plant chance constraint optimization scheduling method based on generalized Nash bargaining

Abstract

The invention discloses a multi-virtual power plant opportunity constraint optimization scheduling method based on generalized Nash bargaining, and the method comprises the steps: 1, considering the P2P transaction of each virtual power plant, constructing a multi-virtual power plant cooperative game model with the minimum total cost as a target based on the cost of each virtual power plant, and building opportunity constraint; 2, establishing a profit distribution model based on generalized Nash bargaining by adopting a generalized Nash bargaining method based on a comprehensive contribution degree and combining with the multi-virtual power plant cooperative game model obtained in the step 1; and step 3, according to the solution result of the profit distribution model based on the generalized Nash bargaining established in the step 2, completing optimal scheduling of the corresponding virtual power plant, and realizing reasonable distribution of cooperation residues. According to the invention, a multi-virtual power plant cooperation transaction strategy considering economy and reliability can be provided, and the cooperation enthusiasm of the virtual power plants is improved.

Description

Technical Field

[0001] The present invention relates to the field of virtual power plant optimization scheduling methods, and in particular to a multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining. Background Art

[0002] As the proportion of new energy continues to increase, the supply and demand balance of the power grid has become increasingly complex and uncertain. Virtual power plants integrate distributed power sources, energy storage and controllable load resources to effectively regulate power supply and demand, thereby improving the flexibility and stability of the power system. Multi-virtual power plant optimization has significant advantages in energy interaction compared to a single virtual power plant. By connecting multiple virtual power plants, a broader energy network can be formed, more efficient energy allocation and utilization can be achieved, the flexibility and reliability of energy interaction can be improved, new energy generation and power demand can be better matched, and enterprise operating costs can be reduced.

[0003] Virtual power plants in adjacent regions can share energy by forming cooperative alliances, reduce the amount of transactions with the upper-level power grid, promote the consumption of local new energy, and thus reduce energy costs. Therefore, forming an alliance is crucial to the realization of energy sharing of virtual power plants, and profit distribution is the key factor in whether the alliance is established.

[0004] There are many studies on profit distribution in cooperative games. Nash bargaining is a common solution to the problem of profit distribution in cooperative games. The basic idea of ​​Nash bargaining is to assume that there is a possibility of cooperation among participants, and they can reach a common distribution plan through negotiation. However, the Nash bargaining method does not take into account the bargaining power of the participants, so that each participant will obtain the same "cost reduction" compared to the negotiation breakdown point. The generalized Nash bargaining takes into account the bargaining power and interests of all parties on the basis of the Nash bargaining method, and further analyzes the profit distribution of participants under different bargaining powers.

[0005] In addition, affected by external environmental factors such as sunlight and weather, the output of renewable energy such as photovoltaics within virtual power plants is uncertain and difficult to accurately predict, which brings risks to the operation of virtual power plants to a certain extent and affects market trading strategies. Therefore, how to conduct in-depth research on the cooperative game mechanism between multiple virtual power plants in an environment with uncertainty in photovoltaic power generation to optimize energy allocation and improve system stability has become an urgent problem to be solved. Summary of the invention

[0006] The present invention provides a multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining to solve the problems of reasonable revenue distribution of multi-virtual power plants and photovoltaic uncertainty in the prior art.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] The multi-virtual power plant opportunity-constrained optimal scheduling method based on generalized Nash bargaining includes the following steps:

[0009] Step 1, considering the P2P transactions of each virtual power plant, based on the energy storage operation cost, fuel cell power generation cost, power transaction cost with the upper power grid, network access fee, backup cost, and carbon trading cost of each virtual power plant, a multi-virtual power plant cooperative game model with the goal of minimizing the total cost is constructed, and the opportunity constraints of the multi-virtual power plant cooperative game model are established, and the constraints include energy storage operation constraints, fuel cell operation constraints, power purchase and sale transaction constraints, carbon quota transaction constraints, photovoltaic system constraints, power balance constraints, distribution network operation constraints, and backup constraints;

[0010] Step 2: Using the generalized Nash bargaining method based on comprehensive contribution, combined with the multi-virtual power plant cooperative game model obtained in step 1, a profit distribution model based on generalized Nash bargaining is established;

[0011] Step 3: Call the Gurobi solver to solve the multi-virtual power plant cooperative game model established in step 1 to obtain the equipment optimization scheduling results of each virtual power plant, and then call the Ipopt solver to obtain the profit distribution results of each virtual power plant based on the generalized Nash bargaining profit distribution model established in step 2 to achieve a reasonable distribution of the cooperative surplus.

[0012] Furthermore, in step 1, a chance constrained optimization method is used to transform the energy storage and fuel cell backup constraints into deterministic constraints based on the impact of PV output forecast deviation on the optimization results of the virtual power plant.

[0013] Furthermore, the comprehensive contribution described in step 2 includes energy contribution and marginal contribution.

[0014] Nash bargaining can ensure that a fair and mutually satisfactory profit distribution plan is reached in the cooperative game, and provides an effective solution to the problem of profit distribution in actual cooperation. On the basis of the Nash bargaining method, generalized Nash bargaining takes into account the bargaining power and interests of all parties, and further analyzes the profit distribution of participants under different bargaining powers. Therefore, this embodiment achieves the optimal scheduling of virtual power plants by establishing a profit distribution model based on generalized Nash bargaining, thereby achieving a reasonable distribution of cooperative surplus.

[0015] Compared with the prior art, the present invention has the following advantages:

[0016] The present invention can reduce the amount of electricity traded between virtual power plants and upper-level power grids through P2P transactions, making the transaction method more flexible and significantly improving economic efficiency. The proposed generalized Nash bargaining method based on comprehensive contribution can measure the impact of the marginal contribution of virtual power plants and the proportion of shared electricity on bargaining power, which is conducive to improving the cooperation enthusiasm of virtual power plants. The opportunity constrained optimization method is used to deal with the impact of photovoltaic uncertainty. The backup needs of virtual power plants can be met through the cooperation of energy storage and fuel cells. At the same time, the virtual power plant can choose an appropriate confidence level during actual operation to balance system reliability and economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining provided in an embodiment of the present invention.

[0018] Figure 2 Schematic diagram of the structure of a virtual power plant system according to an embodiment of the present invention.

[0019] Figure 3 Schematic diagram of photovoltaic prediction and load of a virtual power plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0021] like Figure 1 As shown, this embodiment discloses a multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining, including the following steps:

[0022] Step 1: Consider the P2P transactions of each virtual power plant. Based on the energy storage operation cost, fuel cell power generation cost, power transaction cost with the upper grid, transmission fee, standby cost, and carbon trading cost of each virtual power plant, a multi-virtual power plant cooperative game model with the goal of minimizing the total cost is constructed, as shown in the following formula:

[0023]

[0024]

[0025] Where: T represents the total scheduling period; N i is the number of virtual power plants in the alliance; and They are respectively the energy storage operation cost, fuel cell power generation cost, power transaction cost with the upper-level power grid, transmission fee, standby cost and carbon trading cost of the i-th virtual power plant.

[0026] They represent the energy storage charging and discharging power of the i-th virtual power plant respectively; is the unit operating cost of energy storage; represents the power generation of the fuel cell of the i-th virtual power plant; is the unit electricity generation cost of the fuel cell; They represent the amount of electricity purchased and sold between the ith virtual power plant and the upper-level power grid respectively; Respectively represent the purchase price and sales price of electricity; π i,j represents the electricity transaction cost between building i and building j; represents the P2P transaction volume between the i-th virtual power plant and the j-th virtual power plant; They represent the energy storage reserve capacity and fuel cell reserve capacity of the i-th virtual power plant respectively; are the reserve capacity price of energy storage and the reserve capacity price of fuel cells of the i-th virtual power plant respectively; is the carbon quota trading volume of the i-th virtual power plant; b Base price for purchasing carbon quota; D c is the interval length of the ladder-type carbon trading; ω c is the carbon trading price growth coefficient.

[0027] In addition, this embodiment establishes the opportunity constraints of the multi-virtual power plant cooperative game model, and the opportunity constraints include energy storage operation constraints, fuel cell operation constraints, power purchase and sale transaction constraints, carbon quota transaction constraints, photovoltaic system constraints, power balance constraints, and distribution network operation constraints. The formula is as follows:

[0028] 1) Energy storage operation constraints:

[0029]

[0030]

[0031] Where: Respectively represent the maximum charging power and maximum discharging power of energy storage; Indicates the energy storage capacity level; They represent the charging efficiency and discharging efficiency of energy storage respectively; are the minimum and maximum values ​​of the energy storage capacity respectively.

[0032] 2) Fuel cell operation constraints:

[0033]

[0034] Where: They represent the minimum and maximum power generation of the fuel cell respectively; Represent the upward and downward ramp rates of the fuel cell, respectively.

[0035] 3) Constraints on power purchase and sales transactions:

[0036]

[0037] Where: They are the upper limits of electricity purchase and sales for virtual power plants.

[0038] 4) Carbon quota trading constraints:

[0039]

[0040] Where: is the carbon emission level of the virtual power plant; Free carbon allowances obtained for the allocation of virtual power plants; grid , κ FC They are the unit carbon emission conversion coefficients for electricity purchase and fuel cell power generation respectively.

[0041] 5) Photovoltaic system constraints:

[0042]

[0043] Where: They represent the predicted value and predicted deviation of photovoltaics, is a random quantity, Indicates the actual photovoltaic output.

[0044] 6) Power balance constraints:

[0045]

[0046] Where: Represents the base load of the virtual power plant.

[0047] 7) Distribution network operation constraints:

[0048]

[0049] V n,min ≤V n,t ≤V n,max

[0050] l nm,t ≤l nm,max

[0051] Where: P nm,t , Q nm,t are the active power and reactive power on branch nm respectively; F(n) is the set of end nodes of the branch with node n as the first end node; T(n) is the set of head nodes of the branch with node n as the last end node; l nm,t is the square of the current amplitude of branch nm; V n,t is the square of the voltage amplitude at node n; V n,min 、V n,maxRespectively represent the upper and lower limits of node n; l nm,max is the square upper limit of the current amplitude of branch nm. n,t represents the active power injected into node n, Q n,t represents the reactive power injected by node n, V n,t represents the voltage amplitude of node n, r nm Indicates branch resistance in nm, x nm Represents branch reactance in nm.

[0052] Then, this embodiment adopts the chance constrained optimization method, based on the impact of photovoltaic output forecast deviation on the optimization results of the virtual power plant, to transform the fuel cell and energy storage backup constraints into deterministic constraints, so as to eliminate the impact of photovoltaic output forecast deviation on the optimization results of the virtual power plant and avoid the risk of load loss caused by the uncertainty of photovoltaic output.

[0053] Specifically, considering the existence of photovoltaic output forecast deviation, the virtual power plant reserves a certain amount of backup through energy storage and fuel cells to avoid load loss. At the same time, strict backup constraints may result in high economic costs. Therefore, this embodiment adopts an opportunity constraint optimization method to meet a certain confidence level and take into account the economy and reliability of the system. The backup opportunity constraint optimization method is shown in the following formula:

[0054]

[0055] Where: β represents the confidence level of the system.

[0056] Assuming that the PV output forecast deviation satisfies a Gaussian distribution with an expected value of 0 and a standard deviation of δ, the backup opportunity constraint can be obtained by finding the probability quantile φ -1 (β) is equivalently transformed into a deterministic constraint as shown below:

[0057]

[0058] Therefore, the virtual power plant can select an appropriate confidence level according to the actual operating conditions, provide backup through energy storage and fuel cells, and make timely and flexible adjustments to avoid the risk of load loss caused by the uncertainty of photovoltaic output.

[0059] Step 2: Adopt the generalized Nash bargaining method based on comprehensive contribution, combine it with the multi-virtual power plant cooperative game model obtained in step 1, and establish a profit distribution model based on generalized Nash bargaining.

[0060] In this embodiment, considering that each virtual power plant has the same bargaining power in the Nash bargaining method, the cost of the virtual power plant after allocation is the same as the cost when it is operated independently. However, in fact, the contribution of each virtual power plant is not the same when participating in P2P transactions. Therefore, in order to more reasonably measure the contribution of virtual power plants to the alliance, the generalized Nash bargaining method is used to distribute the profits of virtual power plants, and the bargaining power is quantified from the energy contribution and marginal contribution.

[0061] Energy contribution indicates the proportion of shared electricity of the virtual power plant in P2P transactions, which is expressed as follows:

[0062]

[0063] The marginal contribution represents the contribution of the virtual power plant to the growth of the cooperative alliance's benefits, which is expressed as follows:

[0064]

[0065] Where: v(N i ) represents the benefits of the cooperative alliance; v(N i / i) represents the benefits of the cooperative alliance after removing virtual power plant i.

[0066] The comprehensive contribution The definition is as follows:

[0067]

[0068] The comprehensive contribution has the following characteristics: 1) All virtual power plants participating in P2P transactions will have bargaining power, otherwise the bargaining power is zero and the benefits are also zero. 2) The greater the comprehensive contribution of the virtual power plant, the greater the bargaining power and the greater the benefits.

[0069] In summary, the profit distribution model based on generalized Nash bargaining established in this embodiment is as follows:

[0070]

[0071] Where: and are the operating costs of virtual power plant i in step 1 when it operates independently and participates in P2P transactions, respectively; Represents the profit obtained by virtual power plant i in P2P transactions.

[0072] Step 3: Call the Gurobi solver to solve the multi-virtual power plant cooperative game model established in step 1 to obtain the equipment optimization scheduling results of each virtual power plant, and then call the Ipopt solver to obtain the profit distribution results of each virtual power plant based on the generalized Nash bargaining profit distribution model established in step 2 to achieve a reasonable distribution of the cooperative surplus.

[0073] In this embodiment, a cooperative alliance consisting of three virtual power plants is used as an implementation case to illustrate the effect of the method of this embodiment.

[0074] The system structure of the implementation case of a cooperative alliance consisting of three virtual power plants is as follows Figure 2 As shown in Figure 1, virtual power plants 1, 2 and 3 are connected to nodes 8, 19 and 5 respectively. Each virtual power plant contains energy storage systems, photovoltaics and fuel cells. Figure 3 The upper limit of fuel cell output is set at 100kW, and the up and down climbing capacity is 50kW. The carbon trading base price, carbon emission interval length, and carbon trading penalty growth coefficient are set at 0.1 yuan / kg, 1t, and 1 respectively.

[0075] To verify the effectiveness of the method in this embodiment, the following two transaction modes are set for comparative analysis. Mode 1: The virtual power plant only conducts market transactions with the upper-level power grid and operates independently. Mode 2: The virtual power plant conducts market transactions with the upper-level power grid and conducts P2P transactions internally. Table 1 shows the various costs of the virtual power plant alliance under different transaction modes. Table 1 is as follows:

[0076] Table 1 Comparison of costs under different transaction modes

[0077]

[0078]

[0079] From Table 1, we can see that although the P2P transaction increased the transmission fee by 659.1 yuan in Mode 2, the amount of electricity traded between the virtual power plant and the upper power grid was greatly reduced, and the electricity purchase cost was reduced by 1,898.2 yuan. Accordingly, the carbon emissions of the virtual power plant were significantly reduced, the purchase of carbon trading quotas was greatly reduced, and the carbon trading cost was reduced. Overall, the overall operating cost of the virtual power plant was reduced by 20.1% through P2P trading, which improved the economic efficiency of the virtual power plant.

[0080] In order to verify the effectiveness of the generalized Nash bargaining method for profit distribution proposed in this embodiment, the following three different distribution schemes are set: Scheme 1 is standard Nash bargaining, which does not consider the contribution of each virtual power plant; Scheme 2 is a generalized Nash bargaining method based on energy contribution, which considers the amount of electricity of the virtual power plant in P2P transactions; Scheme 3 is a generalized Nash bargaining method based on comprehensive contribution. The virtual power plant costs after the three schemes are allocated are shown in Table 2. Table 2 is as follows:

[0081] Table 2 Virtual power plant costs under different schemes

[0082]

[0083] As shown in Table 2, the cost of virtual power plants under the three schemes is less than the cost when operating independently (i.e., the initial negotiation value), indicating that the Nash bargaining method can ensure the stability of the virtual power plant alliance. In Scheme 1, the transaction contribution of virtual power plants is not considered, and the bargaining power is 0. Therefore, the cost of virtual power plants after allocation is reduced by 837.7 yuan compared with the initial negotiation value. On the basis of Scheme 1, the energy contribution of virtual power plants in P2P transactions is further considered. The costs of virtual power plants 1 and 3 with higher energy contributions are reduced. Among them, virtual power plant 1 with the highest energy contribution is reduced by 418.8 yuan, while virtual power plant 2 has the lowest energy contribution and the cost increases by 448.2 yuan. On the basis of Scheme 2, the generalized Nash bargaining method based on comprehensive contribution is adopted to further consider the marginal contribution of virtual power plants participating in the alliance. The cost of virtual power plant 1 with higher marginal contribution is further reduced by 106.6 after allocation. On the contrary, the costs of virtual power plants 2 and 3 with lower marginal contributions increase by 97 yuan and 9.5 yuan respectively. It can be seen that the generalized Nash bargaining allocation method based on comprehensive contribution can measure the bargaining power of each virtual power plant from multiple dimensions, ensuring that the income distribution in the cooperative alliance is more fair and reasonable.

[0084] The optimization results of energy storage and fuel cells in virtual power plants are different at different confidence levels, which will affect the operating costs. Table 3 shows the energy storage operating costs, fuel cell power generation costs, power transaction costs, backup costs and total operating costs of the virtual power plant alliance at different confidence levels. Table 3 is as follows:

[0085] Table 3 Cost analysis at different confidence levels

[0086]

[0087] As can be seen from Table 3, as the confidence level increases, the energy storage charging power increases and the fuel cell output power decreases, so the energy storage operating cost and the fuel cell power generation cost increase and decrease respectively. At the same time, the increase in confidence level leads to an increase in backup requirements, and the backup cost of the virtual power plant increases accordingly, and the total cost also gradually increases. In addition, it can be found that when the confidence level is low, the increase in backup cost and total cost is relatively stable. When the confidence level increases from 0.85 to 0.95, the total cost increases by 1.8% and 2.2% respectively, while the total cost increases sharply from 0.95 to 0.99, with an increase of 4.4%. Therefore, the virtual power plant can select the appropriate confidence level according to the actual operating conditions of the system, taking into account reliability and economy, and avoiding high operating costs.

[0088] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. The embodiments described in the present invention are merely descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. The various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction, and such combinations should also be regarded as the contents disclosed in the present disclosure as long as they do not violate the concept of the present invention. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0089] The present invention is not limited to the specific details of the above-mentioned embodiments. Within the technical concept of the present invention and without departing from the design concept of the present invention, various modifications and improvements made to the technical solution of the present invention by technical personnel in this field should fall within the protection scope of the present invention. The technical contents for which protection is sought in the present invention have been fully recorded in the claims.

Claims

1. A multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining, characterized in that: The following steps are involved: Step 1, considering the P2P transactions of each virtual power plant, based on the energy storage operation cost, fuel cell power generation cost, power transaction cost with the upper power grid, network access fee, backup cost, and carbon trading cost of each virtual power plant, a multi-virtual power plant cooperative game model with the goal of minimizing the total cost is constructed, and the opportunity constraints of the multi-virtual power plant cooperative game model are established, and the constraints include energy storage operation constraints, fuel cell operation constraints, power purchase and sale transaction constraints, carbon quota transaction constraints, photovoltaic system constraints, power balance constraints, distribution network operation constraints, and backup constraints; Step 2: adopt the generalized Nash bargaining method based on comprehensive contribution, combine it with the multi-virtual power plant cooperative game model obtained in step 1, and establish a profit distribution model based on generalized Nash bargaining; Step 3: Call the Gurobi solver to solve the multi-virtual power plant cooperative game model established in step 1 to obtain the equipment optimization scheduling results of each virtual power plant, and then call the Ipopt solver to obtain the profit distribution results of each virtual power plant based on the generalized Nash bargaining profit distribution model established in step 2 to achieve a reasonable distribution of the cooperative surplus.

2. The multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining according to claim 1 is characterized in that: In step 1, the opportunity constraint optimization method is used to transform the energy storage and fuel cell backup opportunity constraints into deterministic constraints based on the impact of PV output forecast deviation on the optimization results of the virtual power plant.

3. The multi-virtual power plant opportunity-constrained optimization scheduling method based on generalized Nash bargaining according to claim 1 is characterized in that: The comprehensive contribution described in step 2 includes energy contribution and marginal contribution.

Images