Virtual power plant optimal scheduling method and device of micro-grid, electronic equipment and medium

Abstract

The invention discloses a virtual power plant optimal scheduling method and device of a micro-grid, electronic equipment and a medium, and is used for solving the problem that a traditional virtual power plant optimal scheduling method is difficult to balance interests of multiple parties and is difficult to deal with strong uncertainty of renewable energy output and market price. The method comprises the following steps: acquiring the external cost of the micro-grid, constructing an internal cost optimization model based on the external cost, and solving the internal cost optimization model through mapping to obtain the internal cost and a micro-grid scheduling result; constructing a micro-grid scheduling optimization model according to the internal cost, and solving the micro-grid scheduling optimization model through mapping to obtain a micro-grid scheduling optimization result; and constructing an energy storage scheduling optimization model of the virtual power plant based on the external cost, and solving the energy storage scheduling optimization model through mapping according to the micro-grid scheduling result and the micro-grid scheduling optimization result to obtain an energy storage optimization scheduling result of the virtual power plant.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant scheduling optimization, and in particular to a method, device, electronic equipment and medium for optimizing scheduling of a virtual power plant in a microgrid. Background Art

[0002] With the increasing penetration of renewable energy sources (RES) and the rapid development of the electricity market, microgrids (MGs), as aggregation units of distributed energy resources (DERs), have become a key vehicle for improving energy efficiency and promoting low-carbon transformation. In practical applications, the coordinated operation of multiple microgrids (MMGs) faces various challenges, including uneven distribution of benefits, insufficient privacy protection, and inefficient real-time scheduling.

[0003] Virtual Power Plants (VPPs) aggregate heterogeneous microgrids into a unified entity, enabling coordinated resource scheduling and reducing system costs. However, their efficient management still requires addressing core issues such as internal cost mechanisms, operational constraint satisfaction, and multi-agent decision-making conflicts.

[0004] Existing virtual power plant management frameworks focus primarily on overall profitability, ignoring the autonomy and privacy needs of individual microgrids, making it difficult to balance the interests of multiple parties. Furthermore, traditional optimization methods rely on precise models and global information, making them difficult to cope with the high uncertainty of renewable energy output and market prices. Summary of the Invention

[0005] The present invention provides a method, device, electronic device and medium for optimizing and dispatching a virtual power plant in a microgrid, which are used to solve or partially solve the technical problems that traditional virtual power plant optimization and dispatching methods are difficult to balance the interests of multiple parties and difficult to cope with the strong uncertainty of renewable energy output and market prices.

[0006] The present invention provides a method for optimizing and dispatching a virtual power plant in a microgrid, the method comprising:

[0007] Obtaining the external cost of the microgrid, building an internal cost optimization model based on the external cost, and solving the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results;

[0008] Constructing a microgrid scheduling optimization model according to the internal cost, and solving the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result;

[0009] An energy storage scheduling optimization model of a virtual power plant is constructed based on the external cost, and according to the microgrid scheduling result and the microgrid scheduling optimization result, the energy storage scheduling optimization model is solved by mapping to obtain the energy storage optimization scheduling result of the virtual power plant.

[0010] Optionally, constructing an internal cost optimization model based on the external cost includes:

[0011] An internal cost optimization model is constructed by simultaneously considering the ratio of the current load to the average load in the microgrid, the cumulative electricity price adjustment amount and the external cost; wherein the cumulative electricity price adjustment amount is the cumulative amount of the difference between the external cost and the internal cost.

[0012] Optionally, solving the internal cost optimization model by mapping to obtain internal costs and microgrid scheduling results includes:

[0013] The internal cost optimization model is used as a first microgrid state variable, and according to the input state of the first microgrid state variable, a multi-layer fully connected neural network is used to solve the problem and output a normalized action value of the microgrid;

[0014] constructing an actual internal cost function through action space mapping according to the microgrid normalized action value and the external cost;

[0015] Optimizing and solving the actual internal cost function with the goal of minimizing the internal cost to obtain the internal cost, and constructing a microgrid load cost function based on the internal cost;

[0016] The microgrid load cost function is optimized and solved with the goal of minimizing the microgrid load cost to obtain a microgrid scheduling result.

[0017] Optionally, constructing a microgrid scheduling optimization model according to the internal cost includes:

[0018] At the same time, the ratio of the current load to the average load in the microgrid, the energy storage capacity, the internal cost, the output of the micro-turbine at the last moment, and the accumulated transferred load are considered to construct a microgrid scheduling optimization model.

[0019] Optionally, solving the microgrid scheduling optimization model by mapping to obtain a microgrid scheduling optimization result includes:

[0020] Decomposing the microgrid normalized action value into an initial energy storage normalized action value, an initial gas turbine normalized action value, and an initial transferable load normalized action value;

[0021] The microgrid scheduling optimization model is used as a second microgrid state variable, and according to the input state of the second microgrid state variable, the initial energy storage normalized action value is mapped to the actual energy storage output, the initial gas turbine normalized action value is mapped to the actual gas turbine output, and the initial transferable load normalized action value is mapped to the actual transferable load output through action space mapping;

[0022] Simultaneously considering the operating costs of the actual output of the energy storage, the actual output of the gas turbine, and the actual output of the transferable load, constructing an equipment operating cost function;

[0023] The equipment operation cost function is optimized and solved with the goal of minimizing the equipment operation cost to obtain the microgrid scheduling optimization result.

[0024] Optionally, constructing an energy storage scheduling optimization model for a virtual power plant based on the external cost includes:

[0025] At the same time, the ratio of the current total load to the average total load in the microgrid, the energy storage capacity and the external cost are taken into consideration to construct an energy storage scheduling optimization model for the virtual power plant.

[0026] Optionally, the microgrid scheduling result includes the microgrid output when the microgrid load cost is minimized; the microgrid scheduling optimization result includes the normalized energy storage action value when the equipment operating cost is minimized; and the energy storage scheduling optimization model is solved by mapping based on the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant, including:

[0027] The energy storage scheduling optimization model is used as an energy storage state variable, and based on the input state of the energy storage state variable, an energy storage device output function is constructed through action space mapping according to the microgrid output and the energy storage normalized action value;

[0028] Optimizing and solving the energy storage device output function with the goal of maximizing the energy storage device output to obtain the actual energy storage device output, and constructing the power purchase cost function of the virtual power plant based on the actual energy storage device output;

[0029] The power purchase cost function is optimized and solved with the goal of minimizing the power purchase cost of the virtual power plant to obtain the energy storage optimization scheduling result of the virtual power plant.

[0030] The present invention also provides a virtual power plant optimization scheduling device for a microgrid, comprising:

[0031] An internal cost optimization model processing unit is used to obtain the external cost of the microgrid, build an internal cost optimization model based on the external cost, and solve the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results;

[0032] a microgrid scheduling optimization model processing unit, configured to construct a microgrid scheduling optimization model according to the internal cost, and solve the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result;

[0033] An energy storage scheduling optimization model processing unit is used to construct an energy storage scheduling optimization model of a virtual power plant based on the external cost, and solve the energy storage scheduling optimization model through mapping according to the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant.

[0034] The present invention further provides an electronic device, comprising a processor and a memory:

[0035] The memory is used to store program code and transmit the program code to the processor;

[0036] The processor is used to execute the virtual power plant optimization scheduling method for the microgrid as described in any one of the above items according to the instructions in the program code.

[0037] The present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the virtual power plant optimization scheduling method for the microgrid as described in any one of the above items.

[0038] It can be seen from the above technical solutions that the present invention has the following advantages:

[0039] A method for optimizing the scheduling of virtual power plants in microgrids is provided. First, the external costs of the microgrid are obtained. An internal cost optimization model is constructed based on the external costs. The internal cost optimization model is then solved through mapping to obtain the internal costs and microgrid scheduling results. Next, a microgrid scheduling optimization model is constructed based on the internal costs. This model is then solved through mapping to obtain the microgrid scheduling optimization results. Finally, an energy storage scheduling optimization model for the virtual power plant is constructed based on the external costs. Based on the microgrid scheduling results and the microgrid scheduling optimization results, the energy storage scheduling optimization model is solved through mapping to obtain the virtual power plant energy storage optimization scheduling results. The technical solution of the present invention proposes a hierarchical model with three progressive stages: internal cost management, microgrid optimization scheduling, and virtual power plant energy storage scheduling. Through multi-agent collaborative optimization, this model balances the interests of individual microgrids with the overall benefits of the virtual power plant. Furthermore, through mapping rather than reward function design, the output range of the optimization solution is directly constrained, providing a basis for strictly meeting cross-period cost constraints, thereby significantly reducing computational complexity and default risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A flowchart of the steps of a method for optimizing scheduling of a virtual power plant in a microgrid;

[0042] Figure 2 A schematic diagram of the overall process of a virtual power plant optimization scheduling method for a microgrid;

[0043] Figure 3 This is a structural block diagram of a virtual power plant optimization scheduling device for a microgrid. DETAILED DESCRIPTION

[0044] The embodiments of the present invention provide a method, device, electronic device and medium for optimizing and scheduling a virtual power plant in a microgrid, which are used to solve or partially solve the technical problems that traditional virtual power plant optimization and scheduling methods are difficult to balance the interests of multiple parties and difficult to cope with the strong uncertainty of renewable energy output and market prices.

[0045] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0046] As an example, virtual power plants (VPPs) aggregate heterogeneous microgrids into a unified entity, enabling coordinated resource scheduling and reducing system costs. However, their effective management still requires addressing core issues such as internal cost mechanisms, operational constraint compliance, and multi-agent decision-making conflicts. Existing VPP management frameworks focus primarily on overall profitability, ignoring the autonomy and privacy needs of individual microgrids, making it difficult to balance the interests of multiple stakeholders. Furthermore, traditional optimization methods rely on precise models and global information, making them difficult to cope with the strong uncertainty of renewable energy output and market prices.

[0047] Therefore, one of the core inventive aspects of the present invention is to provide a method for optimizing the scheduling of a virtual power plant that aggregates multiple heterogeneous microgrids. This method proposes a hierarchical DRL (Deep Reinforcement Learning) model with three progressively different stages: internal cost management, microgrid optimization scheduling, and virtual power plant energy storage scheduling. Through multi-agent collaborative optimization, this model balances the interests of individual microgrids with the overall benefits of the virtual power plant. Furthermore, through an action space mapping approach rather than a reward function design, the output range of the DRL optimization solver is directly constrained, providing a basis for strictly meeting cross-period cost constraints. Compared to traditional reward function methods, the action space mapping approach significantly reduces computational complexity and default risk. By adopting the technical solution of this invention, based on a hierarchical DRL energy management framework, virtual power plant operations are decomposed into internal cost accounting, microgrid scheduling, and centralized energy storage control. Multi-agent collaborative optimization is then used to balance individual and global interests, reducing virtual power plant operating costs and microgrid electricity expenditures. This provides a strategy for the collaborative optimization of multi-agent energy systems in highly uncertain environments.

[0048] Reference Figure 1 , shows a flowchart of the steps of a virtual power plant optimization scheduling method for a microgrid provided by an embodiment of the present invention, which may specifically include the following steps:

[0049] Step 101: Obtain the external cost of the microgrid, construct an internal cost optimization model based on the external cost, and solve the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results;

[0050] In a specific implementation, the internal cost optimization model is constructed based on the external cost. The internal cost optimization model can be constructed by simultaneously considering the ratio of the current load to the average load in the microgrid, the cumulative electricity price adjustment amount and the external cost; wherein the cumulative electricity price adjustment amount is the cumulative amount of the difference between the external cost and the internal cost.

[0051] In some embodiments, the execution process of solving the internal cost optimization model by mapping to obtain the internal cost and microgrid scheduling results includes the following sub-steps S01 to S04:

[0052] Step S01: using the internal cost optimization model as the first microgrid state variable, and solving it through a multi-layer fully connected neural network according to the input state of the first microgrid state variable, and outputting a microgrid normalized action value;

[0053] Step S02: constructing an actual internal cost function through action space mapping based on the normalized action value and external cost of the microgrid;

[0054] Step S03: Optimizing and solving the actual internal cost function with the goal of minimizing the internal cost, obtaining the internal cost, and constructing the microgrid load cost function based on the internal cost;

[0055] Step S04: Optimizing and solving the microgrid load cost function with the goal of minimizing the microgrid load cost to obtain a microgrid scheduling result.

[0056] More specifically, an internal cost optimization model (also called an internal cost management optimization model) is established, whose purpose is to dynamically manage internal costs and sales prices based on external costs (mainly external electricity prices) and microgrid demand, and balance the profits of virtual power plants and the motivation of microgrid participation.

[0057] The internal cost optimization model is used as the microgrid state variable. To distinguish it from the microgrid state variable of the microgrid scheduling optimization model in the following steps, it is defined here as the first microgrid state variable. Specifically:

[0058] (1)

[0059] (2)

[0060] Where, represents the first microgrid state variable; Indicates time; and are external and internal costs (electricity prices), respectively; Current load and average load The ratio of Represents the cumulative electricity price adjustment to ensure long-term price constraints.

[0061] According to the input status , using a multi-layer fully connected neural network, the output is the normalized action value of the microgrid (i.e. the normalized microgrid action value) According to the normalized action value of the microgrid and external costs , which is mapped to the actual internal cost function through the following formula:

[0062] (3)

[0063] Where, Represents the price adjustment coefficient, which is used to control the fluctuation range of internal electricity price costs. This parameter can be independently completed by the trained DRL model after setting the constraint parameters before training.

[0064] The above actual internal cost function is optimized with the goal of minimizing internal cost (internal electricity price) to obtain the internal cost , and based on internal costs Construct the microgrid load cost function as shown below:

[0065] (4)

[0066] Where, represents the microgrid load cost function; Represents microgrid exist The load at the moment, i.e. the microgrid output; represents the average daily load of the microgrid; Represents the daily average value of the internal electricity price.

[0067] in, This is the baseline load (not adjustable). For microgrids The actual net load is the baseline load Functions with energy storage, micro turbines, and transferable loads.

[0068] In the microgrid load cost function By subtracting the "average load ×Average electricity price Eliminate the impact of daily fluctuations and focus on optimizing cost strategies. Its practical purpose is to provide a baseline value to eliminate the impact of load and electricity price fluctuations on rewards on different days, making the training process more stable and allowing the model to focus more on the relative optimization of electricity prices rather than absolute fluctuations. While costs are understandably affected by electricity prices, electricity prices are only one of the influencing factors; costs are also affected by factors such as the operation and scheduling strategy.

[0069] By constructing and solving an internal cost optimization model, the virtual power plant can guide the microgrid to reduce electricity purchases during peak hours and increase electricity purchases during off-peak hours by dynamically adjusting electricity prices, thereby smoothing the overall load curve and providing a basis for subsequent energy storage management.

[0070] Step 102: construct a microgrid scheduling optimization model based on the internal cost, and solve the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result;

[0071] In a specific implementation, a microgrid scheduling optimization model is constructed based on internal costs. The microgrid scheduling optimization model can be constructed by simultaneously considering the ratio of the current load to the average load in the microgrid, the energy storage capacity, the internal costs, the output of the micro-turbine at the last moment, and the accumulated transferred load.

[0072] In some embodiments, the execution process of solving the microgrid scheduling optimization model by mapping to obtain the microgrid scheduling optimization result includes the following sub-steps S11 to S14:

[0073] Step S11: decomposing the microgrid normalized action value into an initial energy storage normalized action value, an initial gas turbine normalized action value, and an initial transferable load normalized action value;

[0074] Step S12: using the microgrid dispatch optimization model as the second microgrid state variable, and according to the input state of the second microgrid state variable, mapping the initial energy storage normalized action value to the actual energy storage output, mapping the initial gas turbine normalized action value to the actual gas turbine output, and mapping the initial transferable load normalized action value to the actual transferable load output through action space mapping;

[0075] Step S13: Considering the operating costs of the actual output of the energy storage, the actual output of the gas turbine, and the actual output of the transferable load, an equipment operating cost function is constructed;

[0076] Step S14: Optimizing and solving the equipment operation cost function with the goal of minimizing the equipment operation cost to obtain the microgrid scheduling optimization result.

[0077] More specifically, a microgrid dispatch optimization model is established, the purpose of which is to optimize the operation of local equipment (energy storage, micro turbines, and transferable loads) and minimize operating costs based on the internal electricity prices and costs provided by the virtual power plant.

[0078] The microgrid dispatch optimization model is used as the microgrid state variable. To distinguish it from the microgrid state variable of the internal cost optimization model in the previous step, it is defined here as the second microgrid state variable. Specifically:

[0079] (5)

[0080] Where, represents the second microgrid state variable; Indicates the amount of energy stored; Indicates the output of the micro-turbine (i.e. gas turbine) at the last moment; Indicates the cumulative amount of transferred load.

[0081] The normalized action value of the microgrid is obtained by the following formula: Decomposed into initial energy storage normalized action value , initial gas turbine normalized action value and the initial transferable load normalized action value :

[0082] (6)

[0083] According to the input status , the above normalized action value 、 、 The formula can be used to map the actual output of the corresponding equipment. They are: actual output of energy storage , Actual output of gas turbine , Actual output of transferable load .

[0084] Taking into account the operating costs of the actual output of energy storage, the actual output of the gas turbine, and the actual output of the transferable load, an equipment operating cost function is constructed. The goal of the equipment operating cost function is to minimize negative costs (i.e., minimize equipment operating costs) while compensating for the impact of daily fluctuations. The constructed equipment operating cost function is shown below:

[0085] (7)

[0086] Where, represents the equipment operation cost function; 、 、 They respectively represent the energy storage electricity purchase cost (corresponding to the actual output of energy storage), the gas turbine cost (corresponding to the actual output of the gas turbine) and the transferable load cost (corresponding to the actual output of the transferable load).

[0087] By optimizing and solving the above equipment operation cost function with the goal of minimizing the equipment operation cost, the microgrid scheduling optimization result can be obtained.

[0088] Step 103: construct an energy storage scheduling optimization model for the virtual power plant based on the external cost, and solve the energy storage scheduling optimization model through mapping according to the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant.

[0089] The microgrid dispatch result (i.e. microgrid output) obtained in the above steps ) can be fed back into the virtual grid as input to the energy storage management phase. For example, the microgrid can charge and store energy during low-price periods and discharge it during high-price periods, synergizing with the centralized energy storage of the virtual power plant to further reduce overall electricity purchase costs.

[0090] In a specific implementation, the energy storage scheduling optimization model of the virtual power plant based on external costs can be constructed by simultaneously considering the ratio of the current total load to the average total load in the microgrid, the energy storage capacity and the external costs to construct the energy storage scheduling optimization model of the virtual power plant.

[0091] Combined with the previous content, the microgrid dispatch results include the microgrid output when the microgrid load cost is minimized ; The microgrid dispatch optimization results include the normalized action value of energy storage when the equipment operating cost is minimized .

[0092] In some embodiments, based on the microgrid scheduling results and the microgrid scheduling optimization results, the execution process of solving the energy storage scheduling optimization model by mapping to obtain the energy storage optimization scheduling results of the virtual power plant includes the following sub-steps S21 to S23:

[0093] Step S21: Using the energy storage scheduling optimization model as the energy storage state variable, and based on the input state of the energy storage state variable, according to the microgrid output and the energy storage normalized action value, constructing the energy storage device output function through action space mapping;

[0094] Step S22: Optimizing and solving the energy storage device output function with the goal of maximizing the energy storage device output, obtaining the actual energy storage device output, and constructing the power purchase cost function of the virtual power plant based on the actual energy storage device output;

[0095] Step S23: Optimize and solve the power purchase cost function with the goal of minimizing the power purchase cost of the virtual power plant to obtain the energy storage optimization scheduling result of the virtual power plant.

[0096] More specifically, a storage scheduling optimization model for virtual power plants is established, the purpose of which is to optimize the charging and discharging strategies of centralized energy storage based on the total demand of the microgrid and the external electricity price, and to minimize the electricity purchase cost of the virtual power plant (at this time, the energy storage cost can be ignored).

[0097] The energy storage dispatch optimization model is used as the energy storage state variable, specifically:

[0098] (8)

[0099] (9)

[0100] Where, Represents the energy storage state variable (input state); Indicates total load Average total load ratio.

[0101] According to the input status , microgrid output and energy storage normalized action value It can be mapped to the energy storage device output function through the following formula:

[0102] (10)

[0103] Where, Indicates the output of energy storage equipment; and Respectively represent the energy storage charging and discharging status indicator variables; and Respectively represent the maximum value of energy storage charging and discharging.

[0104] The output function of the energy storage device is optimized to maximize the output of the energy storage device, and the actual output of the energy storage device can be obtained. , and according to the actual energy storage equipment output Construct the electricity purchase cost function of the virtual power plant as shown below:

[0105] (11)

[0106] Where, represents the electricity purchase cost function of the virtual power plant; Indicates the total net load of the microgrid.

[0107] By optimizing and solving this cost function with the goal of minimizing the virtual power plant's electricity purchase costs, we can obtain the optimal energy storage scheduling results for the virtual power plant. By solving this cost function, we can minimize the external market electricity purchase costs while compensating for daily fluctuations.

[0108] By constructing and solving a virtual power plant (VPP) energy storage dispatch optimization model, the VPP energy storage and the microgrid's local energy storage can complement each other. During peak external electricity price periods, the centralized energy storage discharges to reduce electricity purchase costs. During valley periods, the energy storage charges to balance the microgrid's remaining load.

[0109] In an embodiment of the present invention, a method for optimizing and dispatching a virtual power plant that aggregates multiple heterogeneous microgrids is provided. On the one hand, a DRL hierarchical model with three progressive stages, namely internal cost management, microgrid optimization scheduling, and virtual power plant energy storage scheduling, is proposed. Through multi-agent collaborative optimization, the interests of individual microgrids and the overall benefits of the virtual power plant are taken into account. On the other hand, the output range of the DRL optimization solution agent is directly constrained through the action space mapping method rather than the reward function design, providing a basis for strictly meeting the cross-time cost constraints. Compared with the traditional reward function method, the action space mapping solution method can greatly reduce the computational complexity and default risk.

[0110] For better explanation, refer to Figure 2 , which shows a schematic diagram of the overall process of a method for optimizing and dispatching a virtual power plant in a microgrid, provided by an embodiment of the present invention. It should be noted that this embodiment only briefly describes the general process of optimizing and dispatching a virtual power plant in a microgrid. The specific implementation process of each step can be understood by referring to the relevant content in the aforementioned embodiments. A detailed description is omitted here. It is understood that the present invention is not limited to this.

[0111] Step 201: Obtain the external cost of the microgrid, and consider the ratio of the current load to the average load in the microgrid, the accumulated electricity price adjustment, and the external cost to construct an internal cost optimization model;

[0112] Step 202: Solve the internal cost optimization model through action space mapping to obtain internal cost and microgrid scheduling results;

[0113] Step 203: Considering the ratio of the current load to the average load, the energy storage capacity, the internal cost, the output of the micro-turbine at the last moment, and the accumulated transferred load in the micro-grid, a micro-grid scheduling optimization model is constructed;

[0114] Step 204: Solve the microgrid scheduling optimization model through action space mapping to obtain the microgrid scheduling optimization result;

[0115] Step 205: Considering the ratio of the current total load to the average total load in the microgrid, the energy storage capacity, and the external cost, a virtual power plant energy storage scheduling optimization model is constructed;

[0116] Step 206: Based on the microgrid dispatch results and the microgrid dispatch optimization results, the energy storage dispatch optimization model is solved by mapping to obtain the energy storage optimization dispatch results of the virtual power plant.

[0117] Reference Figure 3 , shows a structural block diagram of a virtual power plant optimization and scheduling device for a microgrid provided by an embodiment of the present invention, which may specifically include:

[0118] The internal cost optimization model processing unit 301 is used to obtain the external cost of the microgrid, build an internal cost optimization model based on the external cost, and solve the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results;

[0119] A microgrid scheduling optimization model processing unit 302 is configured to construct a microgrid scheduling optimization model according to the internal cost, and solve the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result;

[0120] The energy storage scheduling optimization model processing unit 303 is used to construct an energy storage scheduling optimization model of the virtual power plant based on the external cost, and solve the energy storage scheduling optimization model through mapping according to the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant.

[0121] In an optional embodiment, the internal cost optimization model processing unit 301 includes:

[0122] An internal cost optimization model construction unit is used to simultaneously consider the ratio of the current load to the average load in the microgrid, the cumulative electricity price adjustment amount and the external cost to construct an internal cost optimization model; wherein the cumulative electricity price adjustment amount is the cumulative amount of the difference between the external cost and the internal cost.

[0123] In an optional embodiment, the internal cost optimization model processing unit 301 includes:

[0124] a microgrid normalized action value output unit, configured to use the internal cost optimization model as a first microgrid state variable, solve the problem through a multi-layer fully connected neural network according to the input state of the first microgrid state variable, and output a microgrid normalized action value;

[0125] an actual internal cost function construction unit, configured to construct an actual internal cost function through action space mapping according to the microgrid normalized action value and the external cost;

[0126] A microgrid load cost function construction unit is used to optimize and solve the actual internal cost function with the goal of minimizing the internal cost, obtain the internal cost, and construct the microgrid load cost function according to the internal cost;

[0127] The microgrid dispatch result solving unit is used to optimize and solve the microgrid load cost function with the goal of minimizing the microgrid load cost to obtain the microgrid dispatch result.

[0128] In an optional embodiment, the microgrid scheduling optimization model processing unit 302 includes:

[0129] The microgrid dispatch optimization model construction unit is used to simultaneously consider the ratio of the current load to the average load in the microgrid, the energy storage capacity, the internal cost, the output of the micro-turbine at the last moment, and the accumulated transferred load to construct a microgrid dispatch optimization model.

[0130] In an optional embodiment, the microgrid scheduling optimization model processing unit 302 includes:

[0131] a normalized value decomposition unit, configured to decompose the microgrid normalized action value into an initial energy storage normalized action value, an initial gas turbine normalized action value, and an initial transferable load normalized action value;

[0132] a normalized value mapping unit, configured to use the microgrid dispatch optimization model as a second microgrid state variable, and according to an input state of the second microgrid state variable, map the initial energy storage normalized action value to the actual energy storage output, map the initial gas turbine normalized action value to the actual gas turbine output, and map the initial transferable load normalized action value to the actual transferable load output through action space mapping;

[0133] an equipment operation cost function construction unit, configured to simultaneously consider the operation costs of the actual output of the energy storage, the actual output of the gas turbine, and the actual output of the transferable load to construct an equipment operation cost function;

[0134] The microgrid scheduling optimization result solving unit is used to optimize and solve the equipment operation cost function with the goal of minimizing the equipment operation cost to obtain the microgrid scheduling optimization result.

[0135] In an optional embodiment, the energy storage scheduling optimization model processing unit 303 includes:

[0136] The energy storage scheduling optimization model construction unit is used to simultaneously consider the ratio of the current total load to the average total load in the microgrid, the energy storage power and the external cost to construct an energy storage scheduling optimization model for the virtual power plant.

[0137] In an optional embodiment, the microgrid scheduling result includes the microgrid output when the microgrid load cost is minimized; the microgrid scheduling optimization result includes the energy storage normalized action value when the equipment operation cost is minimized; the energy storage scheduling optimization model processing unit 303 includes:

[0138] an energy storage device output function construction unit, configured to use the energy storage scheduling optimization model as an energy storage state variable, and construct an energy storage device output function through action space mapping based on an input state of the energy storage state variable, the microgrid output, and the energy storage normalized action value;

[0139] a power purchase cost function construction unit, configured to optimize and solve the energy storage device output function with the goal of maximizing the energy storage device output, obtain the actual energy storage device output, and construct the power purchase cost function of the virtual power plant based on the actual energy storage device output;

[0140] The energy storage optimization scheduling result solving unit is used to optimize and solve the power purchase cost function with the goal of minimizing the power purchase cost of the virtual power plant, and obtain the energy storage optimization scheduling result of the virtual power plant.

[0141] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the aforementioned method embodiment.

[0142] It should be noted that in order to enable those skilled in the art to better distinguish data of the same type but with different actual meanings, some technical features are distinguished by the first and second in the embodiments of the present invention. The first and second are only used for data distinction and have no other special meanings. It can be understood that the present invention does not impose any restrictions on this.

[0143] An embodiment of the present invention further provides an electronic device, the device including a processor and a memory:

[0144] The memory is used to store program codes and transmit the program codes to the processor;

[0145] The processor is used to execute the virtual power plant optimization scheduling method for the microgrid of any embodiment of the present invention according to the instructions in the program code.

[0146] An embodiment of the present invention further provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the virtual power plant optimization scheduling method for a microgrid according to any embodiment of the present invention.

[0147] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0148] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0149] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0150] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0151] 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 the present invention, or the portion 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 causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0152] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A virtual power plant optimization scheduling method for a microgrid, characterized in that: include: Obtaining the external cost of the microgrid, building an internal cost optimization model based on the external cost, and solving the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results; Constructing a microgrid scheduling optimization model according to the internal cost, and solving the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result; An energy storage scheduling optimization model of a virtual power plant is constructed based on the external cost, and according to the microgrid scheduling result and the microgrid scheduling optimization result, the energy storage scheduling optimization model is solved by mapping to obtain the energy storage optimization scheduling result of the virtual power plant.

2. The virtual power plant optimization scheduling method of the microgrid according to claim 1 is characterized in that: The constructing of an internal cost optimization model based on the external cost includes: An internal cost optimization model is constructed by simultaneously considering the ratio of the current load to the average load in the microgrid, the cumulative electricity price adjustment amount and the external cost; wherein the cumulative electricity price adjustment amount is the cumulative amount of the difference between the external cost and the internal cost.

3. The virtual power plant optimization scheduling method of the microgrid according to claim 2 is characterized in that: Solving the internal cost optimization model by mapping to obtain internal costs and microgrid scheduling results includes: The internal cost optimization model is used as a first microgrid state variable, and according to the input state of the first microgrid state variable, a multi-layer fully connected neural network is used to solve the problem and output a normalized action value of the microgrid; constructing an actual internal cost function through action space mapping according to the microgrid normalized action value and the external cost; Optimizing and solving the actual internal cost function with the goal of minimizing the internal cost to obtain the internal cost, and constructing a microgrid load cost function based on the internal cost; The microgrid load cost function is optimized and solved with the goal of minimizing the microgrid load cost to obtain a microgrid scheduling result.

4. The method for optimizing and dispatching a virtual power plant of a microgrid according to claim 3, characterized in that: The constructing of a microgrid dispatch optimization model according to the internal cost includes: At the same time, the ratio of the current load to the average load in the microgrid, the energy storage capacity, the internal cost, the output of the micro-turbine at the last moment, and the accumulated transferred load are considered to construct a microgrid scheduling optimization model.

5. The virtual power plant optimization scheduling method of the microgrid according to claim 4 is characterized in that: Solving the microgrid scheduling optimization model by mapping to obtain a microgrid scheduling optimization result includes: Decomposing the microgrid normalized action value into an initial energy storage normalized action value, an initial gas turbine normalized action value, and an initial transferable load normalized action value; The microgrid scheduling optimization model is used as a second microgrid state variable, and according to the input state of the second microgrid state variable, the initial energy storage normalized action value is mapped to the actual energy storage output, the initial gas turbine normalized action value is mapped to the actual gas turbine output, and the initial transferable load normalized action value is mapped to the actual transferable load output through action space mapping; Simultaneously considering the operating costs of the actual output of the energy storage, the actual output of the gas turbine, and the actual output of the transferable load, constructing an equipment operating cost function; The equipment operation cost function is optimized and solved with the goal of minimizing the equipment operation cost to obtain the microgrid scheduling optimization result.

6. The method for optimizing and dispatching a virtual power plant in a microgrid according to any one of claims 3 to 5, characterized in that: The energy storage scheduling optimization model for the virtual power plant constructed based on the external cost includes: At the same time, the ratio of the current total load to the average total load in the microgrid, the energy storage capacity and the external cost are taken into consideration to construct an energy storage scheduling optimization model for the virtual power plant.

7. The method for optimizing and dispatching a virtual power plant in a microgrid according to claim 6, characterized in that: The microgrid scheduling result includes the microgrid output when the microgrid load cost is minimized; the microgrid scheduling optimization result includes the energy storage normalized action value when the equipment operation cost is minimized; and the energy storage scheduling optimization model is solved by mapping based on the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant, including: The energy storage scheduling optimization model is used as an energy storage state variable, and based on the input state of the energy storage state variable, an energy storage device output function is constructed through action space mapping according to the microgrid output and the energy storage normalized action value; Optimizing and solving the energy storage device output function with the goal of maximizing the energy storage device output to obtain the actual energy storage device output, and constructing the power purchase cost function of the virtual power plant based on the actual energy storage device output; The power purchase cost function is optimized and solved with the goal of minimizing the power purchase cost of the virtual power plant to obtain the energy storage optimization scheduling result of the virtual power plant.

8. A virtual power plant optimization and scheduling device for a microgrid, characterized in that: include: An internal cost optimization model processing unit is used to obtain the external cost of the microgrid, build an internal cost optimization model based on the external cost, and solve the internal cost optimization model through mapping to obtain the internal cost and microgrid scheduling results; a microgrid scheduling optimization model processing unit, configured to construct a microgrid scheduling optimization model according to the internal cost, and solve the microgrid scheduling optimization model through mapping to obtain a microgrid scheduling optimization result; An energy storage scheduling optimization model processing unit is used to construct an energy storage scheduling optimization model of a virtual power plant based on the external cost, and solve the energy storage scheduling optimization model through mapping according to the microgrid scheduling result and the microgrid scheduling optimization result to obtain the energy storage optimization scheduling result of the virtual power plant.

9. 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 optimization scheduling method for the microgrid according to any one of claims 1 to 7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program code, and the program code is used to execute the virtual power plant optimization scheduling method for the microgrid according to any one of claims 1 to 7.

Images