A method for optimizing and scheduling internal resources of a virtual power plant and related devices

Abstract

This application discloses a method and related apparatus for optimizing the scheduling of resources within a virtual power plant. The method includes: constructing a two-stage optimization function for the virtual power plant and determining constraints to obtain an optimization decision model for the virtual power plant. The two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses due to prediction errors. The method then optimizes and solves the virtual power plant optimization decision model to obtain the output results of each distributed resource within the virtual power plant. Finally, it allocates social welfare resources to each distributed resource based on their output results to obtain the social welfare allocation results for each distributed resource. This application achieves rational scheduling of resources within the virtual power plant and improves the stability of the power system.

Description

Technical Field

[0001] This application relates to the field of power system technology, and in particular to a method for optimizing and scheduling internal resources of a virtual power plant and related devices. Background Technology

[0002] Currently, distributed photovoltaic (PV) and distributed wind power have become the main development directions in the new energy field. However, frequent extreme weather events lead to a continuous increase in short-term peak loads, resulting in a growing shortage of flexible adjustment resources required for the safe operation of the power system, and a serious decline in operational reliability and economic efficiency. At the same time, the output of distributed energy resources (DERs) exhibits significant randomness, intermittency, and volatility. Therefore, improving the power system's flexible adjustment capacity to absorb distributed renewable resources has become a critical issue.

[0003] Currently, the main methods to improve the flexibility of the power system are: retrofitting thermal power plants, configuring pumped storage units, gas turbines, and other flexible power sources, as well as adopting energy storage technologies. However, these methods require substantial infrastructure construction and new investment, and have long construction cycles, posing a challenge to achieving the goal of rapidly improving the flexibility of the power system.

[0004] Virtual power plants (VPPs) require only upgrades to the terminals of existing equipment, standing out due to their low cost, high efficiency, and ease of deployment. Through advanced control, networking, information, and metering technologies, VPPs integrate distributed renewable energy sources, incorporating newly discovered adjustable resources—such as renewable resources on the supply side, load-side regulation capabilities, and the flexibility of energy storage—into the power system's dispatch and operation. This effectively improves the power system's flexibility and adaptability.

[0005] However, the internal resource scheduling problem of virtual power plants has become a key factor affecting their continuous operation. In fact, unreasonable scheduling strategies can lead to serious defaults or over-fulfillment by internal members, thus impacting grid stability. Therefore, it is necessary to find a reasonable scheduling method to ensure the long-term effective operation of virtual power plants, making them an important part of improving power system stability. Summary of the Invention

[0006] This application provides a method and related apparatus for optimizing the scheduling of internal resources of a virtual power plant, which is used to rationally schedule internal resources of a virtual power plant and improve the stability of the power system.

[0007] In view of this, the first aspect of this application provides a method for optimizing resource scheduling within a virtual power plant, comprising:

[0008] A two-stage optimization function for a virtual power plant is constructed, and constraints are determined to obtain a virtual power plant optimization decision model. The two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses caused by prediction bias.

[0009] The optimization decision model of the virtual power plant is optimized and solved to obtain the output results of each distributed resource within the virtual power plant;

[0010] Based on the output results of each distributed resource within the virtual power plant, social welfare is allocated to each distributed resource to obtain the social welfare allocation results for each distributed resource.

[0011] Optionally, the process of constructing the day-ahead scheduling optimization function includes:

[0012] The social welfare function of the virtual power plant is constructed based on the total day-ahead dispatched power, the deviation in real-time dispatch caused by the wind and solar power output forecasting error, and the total dispatch cost of the virtual power plant.

[0013] Conditional risk value is introduced into the social welfare function of the virtual power plant to construct a day-ahead scheduling optimization function aimed at maximizing social welfare.

[0014] Optionally, the construction process of the real-time scheduling optimization function includes:

[0015] Based on the deviation in real-time dispatch caused by the wind and solar power output forecasting error and the total dispatching cost of the virtual power plant, a real-time dispatching optimization function is constructed with the objective of minimizing the losses caused by the forecasting error.

[0016] Optionally, the constraints include: total day-ahead bid amount constraints, power balance constraints, gas turbine unit constraints, energy storage equipment constraints, cost constraints, participation status constraints, and conditional risk constraints.

[0017] Optionally, the step of allocating social welfare to each distributed resource based on its output results within the virtual power plant, to obtain the social welfare allocation results for each distributed resource, includes:

[0018] An alliance can be formed based on the participation of each distributed resource within the virtual power plant, or based on the participation of both the distributed resources within the virtual power plant and the virtual power plant operator.

[0019] The total social welfare of the alliance is calculated based on the output results of each distributed resource within the virtual power plant and the electricity price parameters.

[0020] The marginal contribution of each member in the alliance is calculated based on the total social welfare of the alliance, and the weight of each member in the alliance is calculated based on the number of members in the alliance.

[0021] The social welfare distribution of each member is calculated based on their marginal contribution and weight within the alliance.

[0022] The second aspect of this application provides a virtual power plant internal resource optimization and scheduling device, comprising:

[0023] The model building unit is used to construct a two-stage optimization function for a virtual power plant and determine the constraints to obtain a virtual power plant optimization decision model. The two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses caused by prediction bias.

[0024] The solution unit is used to optimize and solve the virtual power plant optimization decision model to obtain the output results of each distributed resource within the virtual power plant.

[0025] The allocation unit is used to allocate social welfare to each distributed resource based on its output results within the virtual power plant, thereby obtaining the social welfare allocation results for each distributed resource.

[0026] Optionally, the process of constructing the day-ahead scheduling optimization function includes:

[0027] The social welfare function of the virtual power plant is constructed based on the total day-ahead dispatched power, the deviation in real-time dispatch caused by the wind and solar power output forecasting error, and the total dispatch cost of the virtual power plant.

[0028] Conditional risk value is introduced into the social welfare function of the virtual power plant to construct a day-ahead scheduling optimization function aimed at maximizing social welfare.

[0029] Optionally, the construction process of the real-time scheduling optimization function includes:

[0030] Based on the deviation in real-time dispatch caused by the wind and solar power output forecasting error and the total dispatching cost of the virtual power plant, a real-time dispatching optimization function is constructed with the objective of minimizing the losses caused by the forecasting error.

[0031] A third aspect of this application provides an electronic device, the device including a processor and a memory;

[0032] The memory is used to store program code and transmit the program code to the processor;

[0033] The processor is used to execute any one of the virtual power plant internal resource optimization scheduling methods described in the first aspect according to the instructions in the program code.

[0034] A fourth aspect of this application provides a computer-readable storage medium for storing program code, which, when executed by a processor, implements the virtual power plant internal resource optimization scheduling method described in any of the first aspects.

[0035] As can be seen from the above technical solutions, this application has the following advantages:

[0036] This application provides a method for optimizing resource scheduling within a virtual power plant, comprising: constructing a two-stage optimization function for the virtual power plant and determining constraints to obtain an optimization decision model for the virtual power plant, wherein the two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses caused by prediction bias; optimizing and solving the optimization decision model for the virtual power plant to obtain the output results of each distributed resource within the virtual power plant; and allocating social welfare for each distributed resource based on its output results to obtain the social welfare allocation results for each distributed resource.

[0037] In this application, a cooperative game theory approach is used to optimize the solution based on the two phases of a virtual power plant, obtaining the output results corresponding to each distributed resource. Cooperation is carried out in both the day-ahead scheduling phase and the real-time scheduling phase to maximize the overall social welfare. At the same time, the social welfare of each distributed resource is allocated according to the total social welfare, ensuring that the social welfare of each member is not lower than the maximum social welfare of its individual participation. This ensures the sustainability of cooperation in the virtual power plant, realizes the rational scheduling of resources within the virtual power plant, and improves the stability of the power system. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A flowchart illustrating a method for optimizing and scheduling internal resources of a virtual power plant, as provided in an embodiment of this application;

[0040] Figure 2 This is a schematic diagram of wind power output scenarios and scenario probabilities provided in the embodiments of this application;

[0041] Figure 3 A schematic diagram of photovoltaic power output scenarios and scenario probabilities provided in the embodiments of this application;

[0042] Figure 4A schematic diagram illustrating the contribution of each member under different risk scenarios, provided in the embodiments of this application;

[0043] Figure 5 A schematic diagram of a virtual power plant internal resource optimization and scheduling device provided in this application embodiment;

[0044] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0046] For easier understanding, please refer to Figure 1 This application provides a method for optimizing resource scheduling within a virtual power plant, comprising:

[0047] Step 101: Construct a two-stage optimization function for the virtual power plant and determine the constraints to obtain the virtual power plant optimization decision model.

[0048] The optimal scheduling of virtual power plants consists of two parts: internal and external. The most critical aspect is how resources are rationally scheduled among the internal members. This application adopts the idea of ​​cooperative game theory to construct an optimal scheduling model based on two-stage cooperative game theory. In the optimal scheduling, Conditional Value at Risk (CVaR) is introduced to measure the price factor risk faced by the Virtual Power Plant (VPP). An optimal decision-making model for virtual power plants is established that meets the requirements of the virtual power plant itself and the power balance constraints of interaction with the external power grid, while also considering the uncertainty risk, in order to obtain the reasonable output result of the internal resources of the virtual power plant.

[0049] For operators, the scheduling and operating costs of each distributed resource are known quantities. During the day-ahead scheduling phase, the optimization function needs to determine the participating day-ahead scheduling strategy while also considering potential losses during the real-time scheduling phase, aiming to maximize social welfare. During the real-time scheduling phase, the optimization function considers minimizing losses due to prediction bias; therefore, there are two functions: a day-ahead scheduling optimization function and a real-time scheduling optimization function. However, in reality, the real-time scheduling optimization function is essentially the same function as the day-ahead scheduling optimization function as predictions become more accurate.

[0050] First, the day-ahead scheduling optimization function for the first stage is determined. In this embodiment, the day-ahead scheduling optimization function adopts an internal scheduling model that considers both risk and cooperative game theory under controllable conditions. Specifically, it can be constructed based on the total day-ahead scheduling power of the virtual power plant, the deviation power of real-time scheduling caused by the wind and solar power output prediction deviation, and the total scheduling cost of the virtual power plant. Conditional risk value is introduced into the social welfare function of the virtual power plant to construct a day-ahead scheduling optimization function with the goal of maximizing social welfare. That is, the day-ahead scheduling optimization function is:

[0051]

[0052] In the formula, β is the risk coefficient of CVaR, 0≤β≤1, and the larger the value of β, the more the virtual power plant seeks lower risk volatility and stable social welfare; p w Let R be the probability of scenario w occurring. wt Let α be the social welfare within the virtual power plant at time t under scenario w, where W is the number of scenarios, T is the scheduling period, α is the confidence level, and η is the VaR value, representing the maximum payoff when the probability that the social welfare of the virtual power plant (VPP) is less than or equal to η is less than or equal to 1-α. CVaR can be expressed as the mathematical expectation that the social welfare of the virtual power plant (VPP) is less than the VaR value η. These are auxiliary variables used when calculating CVaR;

[0053] In this embodiment of the application, the calculation of social welfare takes into account the impact of prices. Assuming that the market manager adopts a dual settlement mechanism for settlement, the following applies:

[0054]

[0055] In the formula, This represents the total dispatched electricity volume of the virtual power plant during the day-ahead phase.

[0056] Secondly, the real-time scheduling optimization function for the second stage is determined. This stage takes place after the day-ahead scheduling clearing, so it is necessary to maximize the social welfare of the real-time scheduling stage, i.e., minimize the overall loss of the virtual power plant caused by the deviation. In the second stage, the output prediction of wind and solar power is on the hourly level and relatively accurate. Therefore, a real-time scheduling optimization function with the objective of minimizing the loss caused by the prediction deviation can be constructed based on the deviation in real-time scheduling caused by the wind and solar power output prediction deviation and the total scheduling cost of the virtual power plant. That is, the real-time scheduling optimization function is:

[0057]

[0058] In the formula, To correct the imbalance in electricity prices, The positive and negative deviations in the real-time dispatch of virtual power plants (VPPs) are due to the existence of wind and solar power output forecasting errors. VPPs need to eliminate these errors in real-time dispatch or purchase corresponding electricity from market regulators. If a VPP's real-time power generation exceeds its day-ahead bid, the excess power generation is charged at a positive imbalance price. Stabilizing and correcting imbalanced electricity prices λ dam The settlement price is for the day-ahead dispatch volume, and it is a 24-hour time-of-use price. If the real-time generation of a VPP is lower than its day-ahead bid volume, its generation deficit will be settled at a negative imbalance price that is greater than or equal to the day-ahead clearing price. Stabilizing and negatively impacting electricity prices C wt This represents the total scheduling cost of the VPP.

[0059] The above formula accurately predicts wind and solar power output. If a multi-period scheduling method is used, the wind and solar power output gradually becomes clear, which is:

[0060]

[0061] Where t slot The two-stage process is a process of continuous adjustment based on the forecast, taking into account the deviation of subsequent time periods while accurately predicting the time points. Although there are differences in form, their essence is the same: to minimize the losses caused by uncertainty.

[0062] Then, the constraints are determined to obtain the virtual power plant optimization decision model. The constraints in this embodiment include day-ahead bid total amount constraints, power balance constraints, gas turbine unit constraints, energy storage equipment constraints, cost constraints, participation status constraints, and conditional risk constraints.

[0063] Day-ahead dispatch constraints: Due to the uncertainty of undispatchable units, it is necessary to determine the dispatchable power of the VPP in the day-ahead phase by predicting the output of each undispatchable member before the day-ahead dispatch quantity is determined, that is:

[0064]

[0065] In the formula, For the output prediction of unschedulable units, r is the number of the unschedulable unit; R is the number of unschedulable units.

[0066] Power balance constraints: The VPP can optimize the output of dispatchable units and balance the dispatched power in real time. The VPP should maintain a balance between power supply and demand in every time period, that is:

[0067]

[0068] In the formula, P represents the actual output of the unschedulable unit r. tgt Let G be the actual output of heat engine unit g, and G be the number of heat engine units. The actual discharge power of the energy storage device s. Let S be the charging power of energy storage device s, and S be the number of energy storage devices.

[0069] Constraints of Gas Turbine Units: The constraints of gas turbine units consist of two parts: power generation cost constraints and physical constraints. The power generation cost constraints are as follows:

[0070] C tgt =a g P tgt +b g S tgt +s g O stgt +d g O dtgt

[0071] In the formula, C tgt For the operating cost of heat unit g, a g b is the marginal cost of the gas turbine unit. g For the fixed costs of gas turbine units, S tgt The state variable is a Boolean variable, which can only be 0 or 1; s g For startup costs, d g For shutdown cost, corresponding to O stgt O dtgt All are Boolean variables, corresponding to the power-on and power-off states, and are either 0 or 1.

[0072] The physical constraints of gas turbine units include:

[0073] Unit output constraints:

[0074] S tgt P gmin ≤P tgt ≤S tgt P gmax

[0075] Startup and shutdown time constraints:

[0076] S tgt -S tg(t-1) -S tgk ≤0

[0077] 1δk-(t-1)≤G g

[0078] S tg(t-1) -S tgt +S tgk ≤1

[0079] 1≤k-(t-1)≤H g

[0080] Start-up and shutdown action constraints:

[0081] -O dtgt ≤S tgt -S tg(t-1) ≤O stgt

[0082] Unit uphill and downhill ramp constraints:

[0083] P tgt -P tg(t-1) ≤(2-S tg(t-1) -S tgt )P gmin +(1+S tg(t-1) -S tgt )UR g P tg(t-1) -S tg ≤(2-S tg(t-1) -S tgt )P gmin +(1-S tg(t-1) +S tgt )DR g

[0084] In the formula, P gmin P represents the lower limit of the output of the gas turbine unit g. gmax G represents the upper limit of the output of the gas turbine unit g; g H is the minimum start-up time for the gas turbine unit g; g UR is the minimum downtime of the gas turbine unit g; g DR is the maximum uphill ramp rate of the gas turbine unit g; g Let g be the maximum downhill ramp rate of the gas turbine unit.

[0085] Constraints of energy storage devices: Constraints of energy storage devices consist of two parts: cost constraints and physical constraints.

[0086] The cost constraint for energy storage equipment is:

[0087]

[0088] In the formula, s is the energy storage device number, a s Let be the marginal operating cost of the energy storage device s;

[0089] Physical constraints on energy storage devices include energy storage constraints, state of charge constraints, and charge / discharge constraints.

[0090] Energy storage energy constraints are:

[0091]

[0092]

[0093]

[0094] In the formula, E swt E sw(t-1) These represent the energy storage states of energy storage device s at time t and time t-1 under scenario w, respectively. The amount of electricity used to charge the energy storage device s; η is the discharge capacity of the energy storage device s; in η is the efficiency constant for energy loss during the charging process of energy storage device s. out Let be the energy loss constant during the discharge process of energy storage device s;

[0095] The state of charge (SOC) constraint is as follows:

[0096]

[0097] SOC smin ≤SOC swt ≤SOC smax

[0098] In the formula, E smax SOC is the maximum energy storage capacity of energy storage devices. smin State of Charge (SOC) is the limit of the state of charge of energy storage devices. smx This represents the upper limit of the state of charge for energy storage devices.

[0099] The charge / discharge constraints are:

[0100]

[0101]

[0102]

[0103] In the formula, This represents the upper limit of the discharge power of the energy storage device s. This represents the upper limit of the charging power of the energy storage device s.

[0104] Cost constraints: Since non-dispatchable units belong to renewable energy units, the generation cost of wind and solar power can be considered zero in day-ahead-real-time dispatch. Therefore, only the cost constraints of energy storage and gas turbine units need to be considered, i.e.:

[0105]

[0106] Participation State Constraints: To maximize social welfare, the marginal contribution of each internal member of the virtual power plant needs to be comprehensively considered. Therefore, this embodiment requires determining the participation state of each member. Participation state fn is defined as follows: If fn = 1, it indicates that the member participates in the virtual power plant; if fn = 0, it indicates that the member does not participate in the virtual power plant, and this member is not considered in the virtual power plant constraints and the maximization of social welfare. The following are the participation constraints for each member:

[0107] The constraints for unschedulable units to participate are:

[0108]

[0109]

[0110] In the formula, fn rwt For unschedulable units, fn rwt =0 means that the unschedulable generating units are not participating in the virtual power plant, fn rwt =1 indicates that unschedulable generating units participate in the virtual power plant, P rmax This is the maximum output limit for non-schedulable generating units;

[0111] The constraints for gas turbine units are:

[0112] fn tgt S tgt P tgmin ≤P tgt ≤fn tgt S tgt P tgmax

[0113] In the formula, fn tgt For, P tgmin For, P tgmax for;

[0114] The constraints for energy storage participation are:

[0115]

[0116]

[0117] In the formula, fn swt fn represents the participation status of energy storage devices. swt =0 means the gas turbine unit did not participate in the virtual power plant, fn swt =1 indicates that energy storage devices participate in the operation of virtual power plants;

[0118] Conditional risk constraints: Conditional risk value is used to characterize the volatility risk posed by unschedulable units, i.e.:

[0119]

[0120]

[0121] By constructing a two-stage optimization function and determining the corresponding constraints, a virtual power plant optimization decision model is obtained.

[0122] The two-stage optimization function constructed in this application operates as follows: In the first stage, the day-ahead stage, pre-scheduling is performed based on the preliminary predicted values ​​of renewable resources, while also considering potential losses due to deviations in the real-time stage. As time progresses, the predictions of renewable resources become more accurate, at which point the second stage of the two-stage optimization function, the real-time stage, begins. In this stage, timely optimization and scheduling adjustments are made based on the accurate predicted values ​​of new energy sources. At this point, the virtual power plant needs to balance power deviations as much as possible to achieve stable operation of the power system. This two-stage optimization function considers potential deviations in the first stage to prepare for actual deviations in the second stage, and then makes timely scheduling adjustments based on the actual deviations in the second stage.

[0123] This two-stage optimization function, on the one hand, reduces the impact of renewable energy resource deviations on the power system by making timely adjustments in advance when renewable energy output decreases; on the other hand, when renewable energy output increases, this situation is considered in the first stage, and the output of fossil fuel units is adjusted accordingly in the second stage, reducing wind and solar curtailment and achieving maximum utilization of renewable resources. By constructing a two-stage optimization function, the risks caused by uncertainty can be reduced, and more rational optimal scheduling can be achieved.

[0124] Step 102: Optimize and solve the virtual power plant optimization decision model to obtain the output results of each distributed resource within the virtual power plant.

[0125] The optimization decision model of the virtual power plant is solved. The decision variables include Boolean variables (used to control the participation status of each member of the virtual power plant) and continuous variables. Therefore, this optimization problem belongs to the category of mixed integer programming and can be solved using the gurobi toolbox of yalmip in Matlab. The output results of each distributed resource within the virtual power plant are obtained. Then, the operators of each distributed resource can schedule according to their corresponding output results.

[0126] Step 103: Based on the output results of each distributed resource within the virtual power plant, perform social welfare allocation on each distributed resource to obtain the social welfare allocation results for each distributed resource.

[0127] In one embodiment, an alliance is formed based on the participation of each distributed resource within the virtual power plant;

[0128] The total social welfare of the alliance is calculated based on the output results of each distributed resource within the virtual power plant and the electricity price parameters.

[0129] The marginal contribution of each member in the alliance is calculated based on the total social welfare of the alliance, and the weight of each member in the alliance is calculated based on the number of members in the alliance.

[0130] The social welfare distribution of each member is calculated based on their marginal contribution and weight within the alliance.

[0131] In this embodiment of the application, the alliance is defined as a set of members (distributed resources within the virtual power plant). Members in the set participate in the optimal scheduling of the virtual power plant, while members outside the set do not participate in the optimal scheduling of the virtual power plant.

[0132] N = {1, 2, 3, ..., n}

[0133] fn i =1 (i∈N)

[0134]

[0135] In the formula, fn i Let i be the participation state vector of member i;

[0136] Similarly, in this embodiment, a sub-alliance is defined as a subset of N, representing a combination of members. Members within a sub-alliance participate in optimized scheduling, while members outside the sub-alliance do not participate in the optimized scheduling of the virtual power plant.

[0137] S∈N

[0138] fn i =1 (i∈S)

[0139]

[0140] Based on the participation state constraints in the virtual power plant optimization decision-making model, the participation state vector f can be obtained. n (S):

[0141] fn(S)=(fn1,fn2,fn3,…,fn n )

[0142] fn i =1 (i∈S)

[0143]

[0144] Therefore, for the participating state vector f n (S), the quantity of which is 2 n -1, that is, 2 n -1 optimization scenario needs to be considered.

[0145] Define the total social welfare of the alliance as v(N), and the corresponding participating state vector as f.n (N); Define the total social welfare of the sub-alliance as v(S), and the corresponding participating state vector as f. n (S). The total social welfare of the alliance can be calculated based on the output results and electricity price parameters of each distributed resource within the virtual power plant. The social welfare of each member in all alliance combinations is determined by calculating their marginal contribution. The marginal contribution of member i is:

[0146] Mc(i S )=v(S)-v(S\i),i s ∈S

[0147] Among them, Mc(i S ) represents the marginal contribution of member i to the sub-alliance S, therefore in f n (S) contains fn i =1, V(S\i) represents fn i When =0, all other elements remain unchanged to form a new sub-alliance S′.

[0148] The social welfare benefits corresponding to each member are as follows:

[0149]

[0150] In the formula, Sw i The social welfare obtained by member i The weight of the marginal contribution of member i to the corresponding sub-alliance S.

[0151] Since all members within a sub-alliance have equal principal status, therefore:

[0152]

[0153] The determination of weights should take into account the overall rationality of the alliance, that is:

[0154]

[0155] The sum of the final social welfare of the members in the formula should be the same as the overall social welfare of the virtual power plant.

[0156] Based on the above constraints, the following can be obtained:

[0157]

[0158] In the formula, |S| is the state constraint vector f n The number of 1s in (S) represents the number of members participating.

[0159] Therefore, without considering virtual power plant operators, the social welfare corresponding to members is:

[0160]

[0161] In another embodiment, an alliance is formed based on the participation of each distributed resource within the virtual power plant and the participation of the virtual power plant operator;

[0162] The total social welfare of the alliance is calculated based on the output results of each distributed resource within the virtual power plant and the electricity price parameters.

[0163] The marginal contribution of each member in the alliance is calculated based on the total social welfare of the alliance, and the weight of each member in the alliance is calculated based on the number of members in the alliance.

[0164] The social welfare distribution of each member is calculated based on their marginal contribution and weight within the alliance.

[0165] The above process does not consider virtual power plant operators, and this social welfare mapping method is not conducive to the sustainable development and operation of virtual power plants. In order to ensure the sustainable development of virtual power plants, virtual power plant operators are considered in the optimal scheduling, and the participation state constraints of virtual power plant operators are defined as follows:

[0166] fn o =0 or 1

[0167] When fn o When fn = 0, the virtual power plant operator does not participate in the virtual power plant optimization and scheduling process. At this time, all members are in a state of perfect competition, with no cooperative relationship, and social welfare is minimized; when fn = 0, the virtual power plant operator does not participate in the virtual power plant optimization and scheduling process. o When the value is 1, the virtual power plant operator participates in the optimization and scheduling process. Each member is organized by the virtual power plant operator to cooperate and participate in the optimization, thereby improving the social welfare corresponding to the virtual power plant.

[0168] Virtual power plant operators participate in optimized dispatching, and in this case, the alliance is defined as:

[0169] N o ={0, 1, 2, 3, ..., n}

[0170] The state constraint vector then becomes:

[0171] fn o (S)=(fn o fn1, fn2, fn3, ..., fn n )

[0172] fn(S)=(fn1,fn2,fn3,…,fn n )

[0173] When fn o When = 0,

[0174] When fn o When = 1, the corresponding participating state vector is fn. o(S), based on the above optimization function and constraints, the optimal objective function value is solved, thus yielding the social welfare of each member:

[0175]

[0176] In the formula, |S| is the state constraint vector fn o The number of 1s in (S) represents the participation status of members and operators;

[0177] The social welfare benefits corresponding to virtual power plant operators are:

[0178]

[0179] When calculating the marginal contribution of members, Mc(i) s When the alliance consists only of the operator and the member, it involves calculating the social welfare v({o}) of the virtual power plant operator alone. Since the virtual power plant operator does not have the conditions to create value without dispatching any distributed energy, the social welfare of the virtual power plant operator is 0, i.e., v({o}) = 0.

[0180] For member i (distributed resource) within the virtual power plant, the corresponding social welfare is:

[0181]

[0182] Among them, Mc(i s The optimized operation of the virtual power plant can guarantee that v(S) - v(S\i) ≥ v({i}), therefore for Mc(i)... S If )≥v({i}), then the final result Sw i It must be greater than the level v({i}) of its individual participation. In the absence of competition from external virtual power plants, it can ensure that members continuously participate in the scheduling optimization of virtual power plants, thus ensuring the sustainability of virtual power plant scheduling and helping to improve the stability of the power system.

[0183] In calculating the marginal contribution Mc(i) of member i S Corresponding weight At that time, adopt This ensures fairness in the social welfare of each member within the virtual power plant, while only considering their marginal contribution Mc(i) during the process. S Anonymity is guaranteed by not considering the specific operational nature of members.

[0184] In this embodiment of the application, the social welfare of the virtual power plant operator is:

[0185]

[0186] Due to the optimization effect of the virtual power plant, we have v(S)-∑ i∈S v({i})≥0, therefore the social welfare of the virtual power plant operator is greater than 0, thus ensuring the sustainable operation of the virtual power plant operator.

[0187] In this process, on the one hand, the virtual power plant operator and its internal resource entities share a common goal: to reduce potential losses and achieve stable power system operation, thus providing a prerequisite for cooperative operation. On the other hand, the virtual power plant's internal resources are complementary; renewable energy units need fossil fuel units to mitigate risks, while fossil fuel units need renewable energy units to achieve green and environmentally friendly operation, demonstrating a willingness to cooperate. Simultaneously, to achieve more stable dispatch, each resource entity within the virtual power plant needs to report information to the virtual power plant operator, who in turn promptly sends dispatch information to each resource entity. This establishes a cooperative relationship between the resource entities and the operator. Each resource entity must cooperate while ensuring its own safety, thus forming a cooperative game theory relationship. Because the internal entities share the same goal, and because cooperative game theory reduces the risks faced by members compared to non-cooperative game theory, centralized dispatch using a cooperative game theory approach is more reasonable and effective.

[0188] In this embodiment, a cooperative game theory approach is used to optimize the solution based on the two stages of the virtual power plant, obtaining the output results corresponding to each distributed resource. Cooperation is carried out in both the day-ahead scheduling stage and the real-time scheduling stage to maximize the overall social welfare. At the same time, the social welfare of each distributed resource is allocated according to the total social welfare, ensuring that the social welfare of each member is not lower than the maximum social welfare of its individual participation. This ensures the sustainability of cooperation in the virtual power plant, realizes the rational scheduling of resources within the virtual power plant, and improves the stability of the power system.

[0189] The above is an embodiment of a virtual power plant internal resource optimization scheduling method provided in this application. The following is a specific application example of a virtual power plant internal resource optimization scheduling method provided in this application.

[0190] This application will consider the acquisition of social welfare and corresponding situations in typical virtual power plant scenarios to demonstrate the effectiveness of the proposed virtual power plant internal resource optimization and scheduling method.

[0191] The main parameters in the embodiments of this application include dispatchable unit parameters and non-dispatchable unit parameters. The dispatchable unit parameters include: gas turbine cost parameters, gas turbine operating parameters and energy storage device operating parameters. Please refer to Tables 1, 2 and 3 for details.

[0192] Table 1 Cost parameters of gas turbine units

[0193] unit <![CDATA[a g (MWh / ￥)]]> <![CDATA[b g (￥)]]> <![CDATA[s g (￥)]]> <![CDATA[d g (￥)]]> Gas turbine unit 1 340 40 80 10

[0194] Table 2 Operating parameters of gas turbine units

[0195] unit <![CDATA[P gmin (MW)]]> <![CDATA[P gmax (MW)]]> <![CDATA[G g (h)]]> <![CDATA[H g (h)]]> <![CDATA[UR g (MW)]]> <![CDATA[DR g (MW)]]> Gas turbine unit 1 2 5.2 1 1 1.8 1.8

[0196] Table 3 Operating parameters of energy storage devices

[0197]

[0198] Regarding the parameters of unschedulable units, due to the influence of uncertainties, this embodiment assumes that the wind and solar random variables follow a historical distribution. First, a Monte Carlo method is used to generate 2000 wind and solar scenarios to describe the uncertainties faced by the virtual power plant. To reduce the variance of the Monte Carlo method and alleviate the computational burden on the model, a probabilistic distance reduction method is used to reduce the generated 2000 scenarios to 10. Wind power output scenarios (including predicted wind power output scenarios and actual wind power output scenarios) and scenario probabilities are as follows: Figure 2 As shown, the photovoltaic power output scenarios (including predicted photovoltaic power output scenarios and actual photovoltaic power output scenarios) and scenario probabilities are as follows: Figure 3 As shown.

[0199] The above parameters are input into the virtual power plant optimization decision model for optimization and solution, so as to obtain the output of the virtual power plant members under different risk conditions and the social welfare situation.

[0200] Considering risk aversion coefficients β of 0.1, 0.5, and 0.9, and a confidence level α of 0.95, the contribution of each member in scenario 1 is as follows: Figure 4 As shown, the operating strategies of dispatchable units such as energy storage and gas turbines differ under different risk aversion levels. This is because as the risk aversion coefficient β increases, the virtual power plant's aversion to risk increases during the dispatching process, leading to corresponding fluctuations.

[0201] Taking into account different risk levels, the expected social welfare of the 10 scenarios was compared with the social welfare obtained by participating alone. The comparison results are shown in Table 4, and the social welfare obtained by the virtual power plant operator is shown in Table 5.

[0202] Table 4 Comparison Results

[0203]

[0204] Table 5 Social Welfare of Virtual Power Plant Operators

[0205] β Social welfare corresponding to virtual power plant operators 0.1 188 0.5 210 0.9 233

[0206] Virtual power plants (VPPs) are essentially unified scheduling systems operated by the VPP itself. VPP operators need to ensure the sustainability of cooperation among their members. Without considering external competition, the VPP must guarantee that the resources it provides to its members are no less than what each member could individually obtain. However, current technologies cannot maintain the sustainability of VPP cooperation. Table 4 shows that in this embodiment, after the operator collectively optimizes the scheduling of the VPP, the social welfare of each member increases compared to when they are scheduled individually. This is because cooperative game theory occurs in both the day-ahead scheduling phase and the real-time scheduling phase, thus increasing overall social welfare. Table 5 shows that the VPP operator also obtains a reasonable allocation of social welfare.

[0207] Please refer to Figure 5 This application also provides a virtual power plant internal resource optimization and scheduling device, including:

[0208] The model building unit is used to construct the two-stage optimization function of the virtual power plant and determine the constraints to obtain the virtual power plant optimization decision model. The two-stage optimization function includes a day-ahead scheduling optimization function with the goal of maximizing social welfare and a real-time scheduling optimization function with the goal of minimizing the losses caused by prediction bias.

[0209] The solution unit is used to optimize and solve the virtual power plant optimization decision model to obtain the output results of each distributed resource within the virtual power plant.

[0210] The allocation unit is used to allocate social welfare to each distributed resource based on its output results within the virtual power plant, thereby obtaining the social welfare allocation results for each distributed resource.

[0211] As a further improvement, the current scheduling optimization function is constructed as follows:

[0212] The social welfare function of the virtual power plant is constructed based on the total day-ahead dispatched power, the deviation in real-time dispatch caused by the wind and solar power output forecasting error, and the total dispatch cost of the virtual power plant.

[0213] Conditional risk value is introduced into the social welfare function of the virtual power plant to construct a day-ahead scheduling optimization function aimed at maximizing social welfare.

[0214] As a further improvement, the construction process of the real-time scheduling optimization function includes:

[0215] Based on the deviation in real-time dispatch caused by the wind and solar power output forecasting error and the total dispatching cost of the virtual power plant, a real-time dispatching optimization function is constructed with the objective of minimizing the losses caused by the forecasting error.

[0216] As a further improvement, the allocation unit is specifically used for:

[0217] An alliance can be formed based on the participation of each distributed resource within the virtual power plant, or based on the participation of both the distributed resources within the virtual power plant and the virtual power plant operator.

[0218] The total social welfare of the alliance is calculated based on the output results of each distributed resource within the virtual power plant and the electricity price parameters.

[0219] The marginal contribution of each member in the alliance is calculated based on the total social welfare of the alliance, and the weight of each member in the alliance is calculated based on the number of members in the alliance.

[0220] The social welfare distribution of each member is calculated based on their marginal contribution and weight within the alliance.

[0221] Please refer to Figure 6 This application also provides an electronic device, which includes a processor and a memory;

[0222] The memory is used to store program code and transfer the program code to the processor;

[0223] The processor is used to execute the virtual power plant internal resource optimization scheduling method in the foregoing method embodiments according to the instructions in the program code.

[0224] This application also provides a computer-readable storage medium for storing program code, which, when executed by a processor, implements the virtual power plant internal resource optimization scheduling method in the aforementioned method embodiments.

[0225] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0226] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0227] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0228] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0229] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0230] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0231] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of this application through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0232] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for optimizing resource scheduling within a virtual power plant, characterized in that, include: A two-stage optimization function for a virtual power plant is constructed, and constraints are determined to obtain a virtual power plant optimization decision model. The two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses due to prediction bias. The day-ahead scheduling optimization function is as follows: ； In the formula, p represents the risk coefficient of CVaR. w Let R be the probability of scenario w occurring. wt Let W represent the social welfare within the virtual power plant at time t under scenario w, where W is the number of scenarios and T is the scheduling period. Confidence level; The VaR value indicates that the social welfare of the virtual power plant is less than or equal to... The probability is less than or equal to 1- The maximum benefit at that time is represented by CVaR, which is the social welfare of the virtual power plant less than the VaR value. The mathematical expectation, These are auxiliary variables used when calculating CVaR; The real-time scheduling optimization function is: ； In the formula, , These are positive imbalance electricity prices and negative imbalance electricity prices, respectively. , These represent the positive and negative deviations in the real-time dispatch of the virtual power plant; C wt The total scheduling cost of the virtual power plant; The constraints include: total day-ahead bid volume constraints, power balance constraints, gas turbine unit constraints, energy storage equipment constraints, cost constraints, participation status constraints, and conditional risk constraints. The optimization decision model of the virtual power plant is optimized and solved to obtain the output results of each distributed resource within the virtual power plant; Based on the output results of each distributed resource within the virtual power plant, social welfare is allocated to each distributed resource to obtain the social welfare allocation results for each distributed resource.

2. The method for optimizing and scheduling internal resources of a virtual power plant according to claim 1, characterized in that, The process of allocating social welfare to each distributed resource based on its output results within the virtual power plant, to obtain the social welfare allocation results for each distributed resource, includes: An alliance can be formed based on the participation of each distributed resource within the virtual power plant, or based on the participation of both the distributed resources within the virtual power plant and the virtual power plant operator. The total social welfare of the alliance is calculated based on the output results of each distributed resource within the virtual power plant and the electricity price parameters. The marginal contribution of each member in the alliance is calculated based on the total social welfare of the alliance, and the weight of each member in the alliance is calculated based on the number of members in the alliance. The social welfare distribution of each member is calculated based on their marginal contribution and weight within the alliance.

3. A virtual power plant internal resource optimization and scheduling device, characterized in that, include: The model building unit is used to construct a two-stage optimization function for a virtual power plant and determine the constraints to obtain a virtual power plant optimization decision model. The two-stage optimization function includes a day-ahead scheduling optimization function aimed at maximizing social welfare and a real-time scheduling optimization function aimed at minimizing losses caused by prediction bias. The day-ahead scheduling optimization function is as follows: ； In the formula, p represents the risk coefficient of CVaR. w Let R be the probability of scenario w occurring. wt Let W represent the social welfare within the virtual power plant at time t under scenario w, where W is the number of scenarios and T is the scheduling period. Confidence level; The VaR value indicates that the social welfare of the virtual power plant is less than or equal to... The probability is less than or equal to 1- The maximum benefit at that time is represented by CVaR, which is the social welfare of the virtual power plant less than the VaR value. The mathematical expectation, These are auxiliary variables used when calculating CVaR; The real-time scheduling optimization function is: ； In the formula, , These are positive imbalance electricity prices and negative imbalance electricity prices, respectively. , These represent the positive and negative deviations in the real-time dispatch of the virtual power plant; C wt The total scheduling cost of the virtual power plant; The constraints include: total day-ahead bid volume constraints, power balance constraints, gas turbine unit constraints, energy storage equipment constraints, cost constraints, participation status constraints, and conditional risk constraints. The solution unit is used to optimize and solve the virtual power plant optimization decision model to obtain the output results of each distributed resource within the virtual power plant. The allocation unit is used to allocate social welfare to each distributed resource based on its output results within the virtual power plant, thereby obtaining the social welfare allocation results for each distributed resource.

4. An electronic device, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the virtual power plant internal resource optimization scheduling method according to any one of claims 1-2 according to the instructions in the program code.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code, which, when executed by a processor, implements the virtual power plant internal resource optimization scheduling method according to any one of claims 1-2.

Images