Virtual power plant active real-time scheduling equivalent model construction method and active scheduling method

Abstract

The invention discloses a virtual power plant active real-time scheduling equivalent model construction method and an active scheduling method. The virtual power plant active real-time scheduling equivalent model construction method comprises the following steps: obtaining a plurality of training active real-time scheduling instructions; obtaining an internal active real-time optimization scheduling model of the virtual power plant; determining a plurality of corresponding training real-time scheduling costs and a plurality of training scheduling instruction feasibility analysis results based on the virtual power plant internal active real-time optimization scheduling model and the plurality of training active real-time scheduling instructions; based on the multiple training active real-time scheduling instructions and the corresponding multiple training real-time scheduling costs, training the first neural network to obtain an active real-time scheduling cost model; and training a second neural network based on the plurality of training active real-time scheduling instructions and the corresponding plurality of training scheduling instruction feasibility analysis results to obtain an active real-time scheduling instruction feasibility model.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a method for constructing an equivalent model of active real-time scheduling of a virtual power plant and an active scheduling method. Background Art

[0002] To minimize the cost of real-time dispatch of active power for a given grid, priority should be given to virtual power plants with low real-time dispatch costs. Therefore, given that each virtual power plant's day-ahead power generation plan has been established, determining the real-time dispatch cost of each virtual power plant for a given real-time dispatch of active power is a pressing issue. Summary of the Invention

[0003] In view of this, an embodiment of the present application provides a method for constructing an equivalent model of real-time active power scheduling of a virtual power plant and an active power scheduling method.

[0004] According to the first aspect of the present application, an embodiment of the present application provides a method for constructing an equivalent model for real-time active power scheduling of a virtual power plant, comprising: A plurality of real-time training active scheduling instructions are obtained, wherein the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power; Obtaining a real-time optimization scheduling model for active power within the virtual power plant; the real-time optimization scheduling model for active power within the virtual power plant satisfies the requirements of training real-time active power scheduling instructions and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost within a scheduling cycle of the virtual power plant; Based on the real-time active power optimization scheduling model within the virtual power plant and multiple training real-time active power scheduling instructions, determine the corresponding multiple training real-time scheduling costs and feasibility analysis results of multiple training scheduling instructions; Based on a plurality of training active real-time scheduling instructions and a plurality of corresponding training real-time scheduling costs, the first neural network is trained to obtain an active real-time scheduling cost model; Based on multiple trained active real-time dispatching instructions and the corresponding multiple training dispatching instruction feasibility analysis results, the second neural network is trained to obtain an active real-time dispatching instruction feasibility model; based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model, the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instructions of the power grid are determined.

[0005] Optionally, a real-time optimization scheduling model for active power within the virtual power plant is obtained, including: Obtaining an objective function corresponding to the real-time optimization scheduling model for active power within the virtual power plant; the objective function is related to a first operating cost function of the thermal power source within the virtual power plant within a scheduling cycle and a second operating cost function of the energy storage system power source within a scheduling cycle; Determine the constraints corresponding to the objective function; the constraints include a first operating constraint corresponding to the thermal power source, a second operating constraint corresponding to the energy storage system power source, and a power balance constraint within the virtual power plant.

[0006] Optionally, the power balance constraints within the virtual power plant include: During the training scheduling period, the sum of the output power of the thermal power supply inside the virtual power plant, the output power of the energy storage system power supply, and the training scheduling power corresponding to the training active real-time scheduling instructions is equal to the total power of the internal load of the virtual power plant.

[0007] Optionally, multiple training merit real-time scheduling instructions are obtained, including: Determine the scheduling start time range, scheduling duration range and scheduling power range corresponding to the real-time scheduling instruction for training active power; Based on the scheduling start time range, scheduling duration range and scheduling power range, a uniform distribution function is used to randomly generate multiple training active real-time scheduling instructions.

[0008] Optionally, based on the real-time active power optimization scheduling model within the virtual power plant and a plurality of training real-time active power scheduling instructions, corresponding multiple training real-time scheduling costs and feasibility analysis results of multiple training scheduling instructions are determined, including: Based on each trained active real-time dispatch instruction, the active real-time optimization dispatch model within the virtual power plant is solved to obtain a solution result; the solution result includes the first output power curve of the thermal power source within the virtual power plant during the training dispatch duration, the second output power curve of the energy storage system power source during the training dispatch duration, and the minimum real-time dispatch cost of the virtual power plant training within a corresponding dispatch cycle; If the solution result shows that the first output power curve, the second output power curve and the minimum training real-time scheduling cost have a solution, the minimum training real-time scheduling cost is used as the corresponding training real-time scheduling cost; and the feasibility analysis result of the training active scheduling instruction is determined to be feasible; If the solution result shows that there is no solution for the first output power curve, the second output power curve or the training minimum real-time scheduling cost, the feasibility analysis result of the training active scheduling instruction is determined to be infeasible.

[0009] According to the second aspect of the present application, an embodiment of the present application provides a method for active power scheduling of a virtual power plant, including: Obtain the active power dispatching instructions from the power grid to the virtual power plant, which include the dispatching start period, dispatching duration period and dispatching power; Based on the active real-time dispatch cost model and active real-time dispatch instruction feasibility model corresponding to each virtual power plant, the active dispatch instruction is processed respectively to obtain the real-time dispatch cost and dispatch instruction feasibility analysis results when each virtual power plant executes the active dispatch instruction; the active real-time dispatch cost model and the active real-time dispatch instruction feasibility model are constructed using the virtual power plant active real-time dispatch equivalent model construction method as in the first aspect or any embodiment of the first aspect; Based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis results of the dispatching instructions, the target virtual power plant is determined, and the target virtual power plant is used to execute the active power dispatching instructions.

[0010] Optionally, based on the real-time dispatch costs corresponding to each virtual power plant and the feasibility analysis results of the dispatch instructions, determining the target virtual power plant includes: Based on the real-time dispatching costs and dispatching instruction feasibility analysis results corresponding to each virtual power plant, the virtual power plant with the feasible dispatching instruction feasibility analysis result and the lowest real-time dispatching cost is determined as the target virtual power plant.

[0011] According to a third aspect of the present application, an embodiment of the present application provides a device for constructing an equivalent model for real-time active power scheduling of a virtual power plant, comprising: The first acquisition module is used to acquire a plurality of real-time training active scheduling instructions, wherein the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power; The second acquisition module is used to obtain the real-time optimization scheduling model of the active power within the virtual power plant; the real-time optimization scheduling model of the active power within the virtual power plant satisfies the requirements of training the real-time active power scheduling instructions and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost of the virtual power plant within a scheduling cycle; A first determination module is used to determine corresponding multiple training real-time scheduling costs and multiple training scheduling instruction feasibility analysis results based on the real-time active power optimization scheduling model within the virtual power plant and multiple training active power real-time scheduling instructions; A first training module is configured to train a first neural network based on a plurality of training active real-time scheduling instructions and a corresponding plurality of training real-time scheduling costs to obtain an active real-time scheduling cost model; The second training module is used to train the second neural network based on multiple trained active real-time dispatching instructions and the corresponding multiple training dispatching instruction feasibility analysis results to obtain an active real-time dispatching instruction feasibility model; based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model, determine the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instructions of the power grid.

[0012] According to a fourth aspect of the present application, an embodiment of the present application provides an active power scheduling device for a virtual power plant, including: The third acquisition module is used to obtain the active power dispatch instruction of the power grid to the virtual power plant, and the active power dispatch instruction includes the dispatch start period, the dispatch duration period and the dispatch power; a processing module for processing active power dispatch instructions based on the active power real-time dispatch cost model and active power real-time dispatch instruction feasibility model corresponding to each virtual power plant, respectively, to obtain real-time dispatch cost and dispatch instruction feasibility analysis results when each virtual power plant executes the active power dispatch instruction; the active power real-time dispatch cost model and the active power real-time dispatch instruction feasibility model are constructed using the virtual power plant active power real-time dispatch equivalent model construction method as described in the first aspect or any embodiment of the first aspect; The second determination module is used to determine the target virtual power plant based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis result of the dispatching instruction, and adopt the target virtual power plant to execute the active power dispatching instruction.

[0013] According to a fifth aspect of the present application, an embodiment of the present application provides an electronic device, including: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so as to enable the at least one processor to execute the method for constructing an equivalent model of real-time active power scheduling of a virtual power plant as in the first aspect or any embodiment of the first aspect, or the method for active power scheduling of a virtual power plant as in the second aspect or any embodiment of the second aspect.

[0014] The embodiment of the present application provides a method for constructing an equivalent model of active real-time scheduling of a virtual power plant and an active scheduling method, which obtains multiple trained active real-time scheduling instructions; obtains an active real-time optimization scheduling model within the virtual power plant; determines multiple corresponding training real-time scheduling costs and multiple feasibility analysis results of the training scheduling instructions based on the active real-time optimization scheduling model within the virtual power plant and the multiple trained active real-time scheduling instructions; trains a first neural network based on the multiple trained active real-time scheduling instructions and the corresponding multiple training real-time scheduling costs to obtain an active real-time scheduling cost model; trains a second neural network based on the multiple trained active real-time scheduling instructions and the corresponding multiple training scheduling instruction feasibility analysis results to obtain an active real-time scheduling instruction feasibility model; in this way, when the virtual power plant's power generation plan has been formulated, the constructed active real-time scheduling cost model and the active real-time scheduling instruction feasibility model can be used to determine the real-time scheduling cost and scheduling instruction feasibility analysis results when the virtual power plant executes the power grid's active real-time scheduling instruction, so that for a given power grid's active real-time scheduling instruction, the active real-time scheduling cost of each virtual power plant can be quickly determined, which facilitates the power grid's active real-time scheduling of the virtual power plant.

[0015] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a method for constructing an equivalent model for real-time active power dispatch of a virtual power plant in an embodiment of the present application; Figure 2 This is a flow chart of a method for active power scheduling of a virtual power plant in an embodiment of the present application; Figure 3 This is a structural diagram of a device for constructing an equivalent model for real-time active power dispatch of a virtual power plant in an embodiment of the present application; Figure 4 This is a structural diagram of an active power dispatching device for a virtual power plant in an embodiment of the present application; Figure 5 This is a schematic diagram of the hardware structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0017] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0018] Because each virtual power plant contains different types of distributed generation (DGs), each with its own privacy protection requirements, each VPP often does not disclose the unit composition, physical model, and operational information of its internal DGs. Therefore, grid dispatch agencies typically only have access to the dispatch capacity and dispatch cost information reported by each VPP based on a given electricity price. Based on this dispatch capacity and dispatch cost information, the grid dispatch agency can determine each VPP's power generation plan within a dispatch cycle. For example, it can determine the planned output power curve for each type of DG within a dispatch cycle, as well as the power curve for power purchased from the external grid. Therefore, given the day-ahead power generation plan of each VPP, to facilitate global real-time active power optimization scheduling by the grid dispatch agency, for a real-time active power dispatch instruction (e.g., [33, 8, 0.76]), which indicates that starting at 8:00 and continuing for two hours, the VPP's power purchase from the external grid should increase by 0.76 MW, each VPP should provide the real-time dispatch cost corresponding to any dispatch power at any time the next day. This is also known as the real-time active power dispatch equivalent model for the VPP.

[0019] To this end, an embodiment of the present application provides a method for constructing an equivalent model for real-time active power scheduling of a virtual power plant, which is applied to a virtual power plant. The method includes: S101: Acquire multiple real-time training active scheduling instructions, where the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power.

[0020] In this embodiment, the virtual power plant may include wind power, hydropower, photovoltaic power, energy storage system power, thermal power, load, etc. Among them, the thermal power may be composed of multiple groups of micro turbines.

[0021] In this embodiment, the equivalent model of the real-time active power dispatch of the virtual power plant is: given the planned output power curves of the various types of distributed power sources in the virtual power plant and the power curves of electricity purchased from the external power grid, a mapping model is established between the real-time active power dispatch instructions for the next day and the real-time dispatch costs within the virtual power plant. Therefore, in order to construct the equivalent model of the real-time active power dispatch of the virtual power plant, a plurality of trained real-time active power dispatch instructions can be obtained first. The trained dispatch power can be the purchased power added from the external power grid during the training dispatch period, or the output power added to the external power grid. When the trained dispatch power is positive, it indicates an increase in the purchased power from the external power grid; when the trained dispatch power is negative, it indicates an increase in the output power to the external power grid.

[0022] S102, obtain the real-time optimization scheduling model of active power within the virtual power plant; the real-time optimization scheduling model of active power within the virtual power plant meets the requirements of training real-time active scheduling instructions, and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost of the virtual power plant within a scheduling cycle.

[0023] In this embodiment, at a given time interval, the real-time dispatch capacity of a virtual power plant is determined by the dispatch capacity of each unit it contains, while the real-time dispatch cost is determined by the marginal cost of regulating a certain power at the current operating point of each unit within it. However, within a certain time period, the regulation behavior occurring at a particular time interval can indirectly affect the dispatch capacity and dispatch cost of other future time intervals. Therefore, the real-time dispatch capacity and dispatch cost of a virtual power plant should not only be considered at the current moment, but should be comprehensively optimized and determined within a scheduling cycle. Therefore, when the virtual power plant's day-ahead power generation plan has been established, for a given real-time active power dispatch instruction from the grid, the virtual power plant should obtain the corresponding dispatch cost by minimizing the real-time dispatch cost of its internal units within a scheduling cycle. Therefore, the virtual power plant can determine the minimum real-time dispatch cost of the virtual power plant within a scheduling cycle by constructing a real-time active power optimization dispatch model within the virtual power plant, which aims to minimize the real-time dispatch cost of the virtual power plant within a scheduling cycle, while meeting the requirements of the trained real-time active power dispatch instruction and determining the planned output power curves of each type of distributed power source within the virtual power plant within a scheduling cycle.

[0024] S103, based on the real-time active power optimization scheduling model within the virtual power plant and multiple training real-time active power scheduling instructions, determine the corresponding multiple training real-time scheduling costs and multiple training scheduling instruction feasibility analysis results.

[0025] In this embodiment, the active real-time optimization scheduling model within the virtual power plant can be solved separately through multiple training active real-time scheduling instructions to obtain corresponding multiple training real-time scheduling costs and multiple training scheduling instruction feasibility analysis results.

[0026] S104: Based on the plurality of training active real-time scheduling instructions and the corresponding plurality of training real-time scheduling costs, the first neural network is trained to obtain an active real-time scheduling cost model.

[0027] In this embodiment, a plurality of training active real-time scheduling instructions may be used as training data and a plurality of training real-time scheduling costs may be used as labels to train the first neural network and obtain an active real-time scheduling cost model.

[0028] During specific implementation, according to the needs of training and testing the virtual power plant active real-time scheduling equivalent model, multiple training active real-time scheduling instructions and corresponding multiple training real-time scheduling costs, and multiple training scheduling instruction feasibility analysis results can be divided into training sets and test sets. ; ; ; ; ; S Ttrain 、D Train 、 、F Ctrain 、F Atrain is the training set, S Ttest 、D Ttest 、 、 F Atest For the test set. S T Schedule the start time for training, D T Schedule duration for training, Power scheduling for training scheduling, F C Real-time scheduling cost for training, F A The feasibility analysis results of training scheduling instructions.

[0029] Then the training dataset S Ttrain 、 D Ttrain 、 is the input of the deep neural network model, F Ctrain For output, the number of layers, number of latent nodes, activation function, optimization solver, and the size of the training data set are determined through experiments. During the experiment, the size of the data set is gradually increased from small to large, and the data set is used after training. S Ttest 、 D Ttest 、 、F Ctest Test until the model accuracy reaches the target accuracy requirement.

[0030] S105. Based on multiple trained active real-time dispatching instructions and corresponding multiple trained dispatching instruction feasibility analysis results, the second neural network is trained to obtain an active real-time dispatching instruction feasibility model; based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model, the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instruction of the power grid are determined.

[0031] In specific implementation, the training data set can be S Ttrain 、 D Ttrain 、 is the input of the classification model, F Atrain For output, a variety of decision tree, support vector machine, nearest neighbor, ensemble learning and other classification models are used for training, and after training, the data set is used S Ttest 、 D Ttest 、 、 F Atest Conduct tests and select the best classification model.

[0032] The embodiment of the present application provides a method for constructing an equivalent model of active real-time dispatch of a virtual power plant, which obtains a plurality of trained active real-time dispatch instructions; obtains an active real-time optimization dispatch model within the virtual power plant; determines a plurality of corresponding training real-time dispatch costs and a plurality of feasibility analysis results of the training dispatch instructions based on the active real-time optimization dispatch model within the virtual power plant and the plurality of trained active real-time dispatch instructions; trains a first neural network based on the plurality of trained active real-time dispatch instructions and the corresponding plurality of training real-time dispatch costs to obtain an active real-time dispatch cost model; trains a second neural network based on the plurality of trained active real-time dispatch instructions and the corresponding plurality of feasibility analysis results of the training dispatch instructions to obtain an active real-time dispatch instruction feasibility model; in this way, when the virtual power plant's power generation plan has been formulated a day ago, the real-time dispatch cost and the feasibility analysis results of the dispatch instructions when the virtual power plant executes the active real-time dispatch instructions of the power grid can be determined by means of the constructed active real-time dispatch cost model and the active real-time dispatch instruction feasibility model, so that for a given power grid active real-time dispatch instruction, the active real-time dispatch cost of each virtual power plant can be quickly determined, which facilitates the power grid's active real-time dispatch of the virtual power plant.

[0033] In an optional embodiment, step S102, obtaining a real-time optimization scheduling model for active power within the virtual power plant, includes: Obtain the objective function corresponding to the real-time optimization scheduling model of active power within the virtual power plant; the objective function is related to the first operating cost function of the thermal power source within the virtual power plant within a scheduling cycle and the second operating cost function of the energy storage system power source within a scheduling cycle; determine the constraints corresponding to the objective function; the constraints include the first operating constraint corresponding to the thermal power source, the second operating constraint corresponding to the energy storage system power source, and the power balance constraint within the virtual power plant.

[0034] In specific implementation, when a virtual power plant's day-ahead power generation plan has been established, the power purchased by the virtual power plant from the external grid is a fixed value, which can be equivalent to a negative load. The wind power, photovoltaic power, and hydropower sources of the virtual power plant are non-adjustable units, which can also be equivalent to a negative load. Therefore, in the real-time optimization of active power within the virtual power plant, only adjustable units such as micro-turbines and energy storage are considered, and the adjustable units respond to the real-time active power dispatch instructions of the grid. Therefore, the objective function of the real-time optimization of active power within the virtual power plant is expressed as follows: ; in, T is the number of time periods included in a scheduling cycle, C MT (t) is the operating cost of the thermal power source during period t, that is, the operating cost of the micro-turbine. C ES (t) It represents the operating cost of the energy storage power supply during period t, that is, the operating cost of the energy storage unit. The calculation formula for each part of the cost is as follows: 1) Operating costs of micro turbines ; in, C MT,m represents the unit power cost of micro-turbine unit m, N MT Indicates the number of micro-turbine units, P MT,m (t) represents the output power of micro-turbine unit m during period t, The length of a period, C MT,m represents the startup cost of micro-turbine unit m, u su,m It represents the startup variable of the micro-turbine unit m in period t, which is 1 when it is started and 0 when it is not started.

[0035] In some embodiments, since the virtual power plant is responding to the real-time active power dispatch instruction of the power grid when a day-ahead power generation plan has been formulated, the output power of the micro-turbine unit in period t can be a fixed value outside the training scheduling period.

[0036] 2) Operating costs of energy storage units , where N ES Indicates the number of energy storage units, C ES,e Indicates the discharge cost of the energy storage unit e when discharging, unit: yuan / kWh; Indicates the discharge power of unit e, unit: kW.

[0037] Due to the energy storage charging power The electricity cost is included in the production cost of other power sources, so the energy storage charging cost is not included here. P ES,e (t) can be expressed as (positive value indicates discharge, negative value indicates charge): .

[0038] In some embodiments, since the virtual power plant is responding to the real-time active power dispatch instruction of the power grid when a day-ahead power generation plan has been formulated, the output power of the energy storage unit in period t can be a fixed value outside the training scheduling period.

[0039] The first operating constraints corresponding to the thermal power supply include: ① Power upper and lower limit constraints: , , in 、 Respectively represent the maximum output and minimum output of unit m, unit: kW; u MT,m (t) is the running variable, which is 1 if running, otherwise 0; β is the safety margin factor.

[0040] ②Climbing rate constraint , , in 、 They represent the upward ramp rate limit and downward ramp rate limit of unit m respectively, in kW / min.

[0041] ③Minimum downtime constraint , in is the minimum downtime of unit m, unit: period; u sd,m (t) It is the shutdown variable, which is 1 if the shutdown occurs and 0 otherwise.

[0042] ④Continuous running time constraints Minimum continuous running time constraint: , in, is the minimum continuous operation time of unit m, in units of time periods. The second operation constraint corresponding to the energy storage system power supply includes: ①Charge and discharge constraints , , in 、 They represent the maximum discharge power and maximum charging power of unit e respectively, 、 Respectively represent the minimum discharge power and minimum charging power of unit e, unit: kW; 、 Represent the discharge state and charge state respectively. When the value is 1, it means discharge / charge, and when the value is 0, it means no discharge / no charge. The two are mutually exclusive, that is, they meet the following constraints: .

[0043] ②SOC constraint , in SOC e (t) It represents the SOC of unit e in period t; SOC emax (t) 、 SOC emin (t) Respectively represent the maximum and minimum allowable values of SOC; SOC e (0) Indicates the initial capacity of SOC; SOC e (t) Indicates the SOC capacity of the last period.

[0044] In some embodiments, the internal power balance constraints of the virtual power plant include: During the training scheduling period, the sum of the output power of the thermal power supply inside the virtual power plant, the output power of the energy storage system power supply, and the training scheduling power corresponding to the training active real-time scheduling instructions is equal to the total power of the internal load of the virtual power plant.

[0045] In specific implementation, if a virtual power plant's day-ahead power generation plan has been established, the power purchased by the virtual power plant from the external grid is a fixed value, which can be equivalent to a negative load. The wind power, photovoltaic power, and hydropower sources of the virtual power plant are non-adjustable units, which can also be equivalent to a negative load. Therefore, in the real-time optimization of active power within the virtual power plant, only adjustable units such as micro-turbines and energy storage are considered, and the adjustable units respond to the real-time active power dispatch instructions of the grid. Therefore, the power balance constraint within the virtual power plant is: ; Where, N MT is the number of micro-turbine units, P MT,m (t) represents the output power of micro-turbine unit m in period t; N ES Indicates the number of energy storage units; P ES,e (t) Indicates the output power of the energy storage unit. A positive value indicates discharge, and a negative value indicates charging. is the real-time dispatching power of the grid during period t. A positive value indicates that the power of the virtual power plant from the external grid is increased. It is the real-time net load total power in the virtual power plant during period t after excluding non-adjustable units and power purchased from the external power grid.

[0046] In this embodiment, since the virtual power plant responds to the real-time active power dispatch instruction of the power grid when the day-ahead power generation plan of the virtual power plant has been formulated, the total power of the load in the virtual power plant during the period t is a fixed value. Therefore, To determine the value.

[0047] In this embodiment, the internal power balance constraint conditions of the virtual power plant are set to include: during the training scheduling period, the sum of the output power of the thermal power source inside the virtual power plant, the output power of the energy storage system power source, and the training scheduling power corresponding to the training active real-time scheduling instructions is equal to the total power of the internal load of the virtual power plant; in this way, only adjustable units such as micro-turbines and energy storage can be considered, and by having the adjustable units respond to the active real-time scheduling instructions of the power grid, the internal power balance constraint conditions of the virtual power plant and the corresponding objective function can be simplified.

[0048] In this embodiment, by obtaining the objective function corresponding to the real-time optimization scheduling model of active power within the virtual power plant and determining the constraints corresponding to the objective function, the real-time optimization scheduling model of active power within the virtual power plant can be quickly constructed, so that the real-time optimization scheduling model of active power within the virtual power plant takes into account both the real-time scheduling cost and the operating constraints of the energy storage system power supply and thermal power supply within the virtual power plant, as well as the scheduling requirements for meeting the active real-time scheduling instructions.

[0049] In an optional embodiment, step S101, obtaining multiple training active real-time scheduling instructions, includes: Determine the scheduling start time range, scheduling duration time range and scheduling power range corresponding to the training active real-time scheduling instruction; based on the scheduling start time range, scheduling duration time range and scheduling power range, use a uniform distribution function to randomly generate multiple training active real-time scheduling instructions.

[0050] In specific implementation, the training schedule starts with the dataset This may include: ; Where, N sample is the size of the dataset; It represents the starting period of the lth sample data. Considering its possible value range, it is randomly generated by the uniform distribution function, that is: ; Where, The value range is The uniform distribution law of The minimum and maximum values allowed for the start period, respectively.

[0051] The training schedule duration dataset may include: ; Where, represents the scheduling duration of the lth sample data. Similarly, according to its possible value range, it is randomly generated using the uniform distribution function, that is: ; Where, These are the minimum and maximum values allowed for the duration period, respectively.

[0052] Training scheduling power datasets may include: ; Where, represents the real-time dispatching power of the lth sample data. Similarly, according to its possible value range, it is randomly generated using a uniform distribution function, that is: ; Where, They are the minimum and maximum values allowed for real-time active dispatch power.

[0053] In this embodiment, by determining the scheduling start time period range, scheduling duration time period range and scheduling power range corresponding to the training active real-time scheduling instruction; based on the scheduling start time period range, scheduling duration time period range and scheduling power range, a uniform distribution function is used to randomly generate multiple training active real-time scheduling instructions; in this way, multiple randomly distributed training active real-time scheduling instructions can be quickly generated.

[0054] In an optional embodiment, step S103, based on the real-time active power optimization scheduling model within the virtual power plant and the multiple training real-time active power scheduling instructions, determines the corresponding multiple training real-time scheduling costs and the feasibility analysis results of the multiple training scheduling instructions, including: Based on each trained active real-time dispatch instruction, the active real-time optimization dispatch model inside the virtual power plant is solved to obtain a solution result; the solution result includes the first output power curve of the thermal power source in the virtual power plant during the training dispatch duration period, the second output power curve of the energy storage system power source during the training dispatch duration period, and the training minimum real-time dispatch cost of the virtual power plant during a corresponding dispatch cycle; if the solution result shows that the first output power curve, the second output power curve and the training minimum real-time dispatch cost have solutions, the training minimum real-time dispatch cost is used as the corresponding training real-time dispatch cost; and the feasibility analysis result of the trained active dispatch instruction is determined to be feasible; if the solution result shows that the first output power curve, the second output power curve or the training minimum real-time dispatch cost have no solutions, the feasibility analysis result of the trained active dispatch instruction is determined to be infeasible.

[0055] In specific implementation, the objective function corresponding to the real-time active power optimization scheduling model within the virtual power plant can be solved based on each training active real-time scheduling instruction and the constraints corresponding to the real-time active power optimization scheduling model within the virtual power plant. If the first output power curve of the thermal power source within the virtual power plant during the training scheduling duration, the second output power curve of the energy storage system power source during the training scheduling duration, and the minimum real-time scheduling cost of the virtual power plant training within a corresponding scheduling cycle can be solved, then the minimum real-time scheduling cost of the training is used as the corresponding real-time scheduling cost, and the feasibility analysis result of the training active power scheduling instruction is determined to be feasible; if the first output power curve of the thermal power source within the virtual power plant during the training scheduling duration, the second output power curve of the energy storage system power source during the training scheduling duration, or the minimum real-time scheduling cost of the virtual power plant training within a corresponding scheduling cycle cannot be solved, then the feasibility analysis result of the training active power scheduling instruction is determined to be infeasible.

[0056] In this embodiment, a plurality of training real-time scheduling costs corresponding to a plurality of training merit real-time scheduling instructions and a plurality of feasibility analysis results of the training scheduling instructions can be quickly determined.

[0057] The embodiment of the present application provides a method for scheduling active power of a virtual power plant, such as Figure 2 Shown, including: S201, obtaining the active power dispatching instruction of the power grid to the virtual power plant, where the active power dispatching instruction includes the dispatching start period, the dispatching duration period and the dispatching power.

[0058] S202, based on the active real-time dispatching cost model and active real-time dispatching instruction feasibility model corresponding to each virtual power plant, the active dispatching instructions are processed respectively to obtain the real-time dispatching cost and dispatching instruction feasibility analysis results when each virtual power plant executes the active dispatching instruction; the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model are constructed using the virtual power plant active real-time dispatching equivalent model construction method in any of the above-mentioned implementation methods.

[0059] S203: Based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis result of the dispatching instruction, a target virtual power plant is determined, and the target virtual power plant is used to execute the active power dispatching instruction.

[0060] The active power dispatching method of the virtual power plant provided in the embodiment of the present application obtains the active power dispatching instruction of the power grid to the virtual power plant, and the active power dispatching instruction includes the dispatching start period, the dispatching duration period and the dispatching power; based on the active power real-time dispatching cost model and the active power real-time dispatching instruction feasibility model corresponding to each virtual power plant, the active power dispatching instruction is processed respectively to obtain the real-time dispatching cost and dispatching instruction feasibility analysis results when each virtual power plant executes the active power dispatching instruction; the active power real-time dispatching cost model and the active power real-time dispatching instruction feasibility model adopt the virtual power plant active power real-time dispatching equivalent model construction method in any of the above-mentioned implementation modes. The method is constructed; based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis results of the dispatching instruction, the target virtual power plant is determined, and the target virtual power plant is used to execute the active power dispatching instruction; in this way, when the virtual power plant's power generation plan has been formulated a day ago, the real-time dispatching cost and the feasibility analysis results of the dispatching instruction when the virtual power plant executes the active power real-time dispatching instruction of the power grid can be determined through the constructed active power real-time dispatching cost model and the active power real-time dispatching instruction feasibility model, so that for a given active power real-time dispatching instruction of the power grid, the active power real-time dispatching cost of each virtual power plant can be quickly determined, which is convenient for the power grid to dispatch the virtual power plant in real-time.

[0061] In an optional embodiment, in step S203, determining a target virtual power plant based on the real-time dispatch costs and dispatch instruction feasibility analysis results corresponding to each virtual power plant includes: Based on the real-time dispatching costs and dispatching instruction feasibility analysis results corresponding to each virtual power plant, the virtual power plant with the feasible dispatching instruction feasibility analysis result and the lowest real-time dispatching cost is determined as the target virtual power plant.

[0062] In this embodiment, by determining the virtual power plant whose feasibility analysis result of the scheduling instruction is feasible and has the lowest real-time scheduling cost as the target virtual power plant, the power grid can minimize the active real-time scheduling cost of the entire network in the active real-time optimization scheduling, thereby saving the cost of the active real-time optimization scheduling of the power grid.

[0063] The embodiment of the present application also provides a device for constructing a virtual power plant active real-time dispatch equivalent model, such as Figure 3 Shown, including: The first acquisition module 31 is configured to acquire a plurality of real-time training active scheduling instructions, wherein the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power.

[0064] The second acquisition module 32 is used to obtain the real-time optimization scheduling model of active power within the virtual power plant; the real-time optimization scheduling model of active power within the virtual power plant meets the requirements of training real-time active scheduling instructions and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost of the virtual power plant within a scheduling cycle.

[0065] The first determination module 33 is used to determine corresponding multiple training real-time scheduling costs and multiple training scheduling instruction feasibility analysis results based on the real-time active power optimization scheduling model within the virtual power plant and multiple training active real-time scheduling instructions.

[0066] The first training module 34 is used to train the first neural network based on a plurality of training active real-time scheduling instructions and a corresponding plurality of training real-time scheduling costs to obtain an active real-time scheduling cost model.

[0067] The second training module 35 is used to train the second neural network based on multiple trained active real-time dispatching instructions and the corresponding multiple training dispatching instruction feasibility analysis results to obtain an active real-time dispatching instruction feasibility model; based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model, to determine the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instruction of the power grid.

[0068] The embodiment of the present application provides a device for constructing an equivalent model of active real-time dispatch of a virtual power plant. When the virtual power plant's power generation plan has been formulated a day ago, the constructed active real-time dispatch cost model and active real-time dispatch instruction feasibility model can be used to determine the real-time dispatch cost and dispatch instruction feasibility analysis results when the virtual power plant executes the active real-time dispatch instruction of the power grid. Therefore, for a given active real-time dispatch instruction of the power grid, the active real-time dispatch cost of each virtual power plant can be quickly determined, which facilitates the active real-time dispatch of the virtual power plant by the power grid.

[0069] The embodiment of the present application also provides a virtual power plant active power scheduling device, such as Figure 4 Shown, including: The third acquisition module 41 is used to obtain the active power dispatching instruction of the power grid to the virtual power plant, and the active power dispatching instruction includes the dispatching start period, the dispatching duration period and the dispatching power.

[0070] The processing module 42 is used to process the active power dispatching instructions based on the active power real-time dispatching cost model and the active power real-time dispatching instruction feasibility model corresponding to each virtual power plant, and obtain the real-time dispatching cost and dispatching instruction feasibility analysis results when each virtual power plant executes the active power dispatching instruction; the active power real-time dispatching cost model and the active power real-time dispatching instruction feasibility model are constructed using the virtual power plant active power real-time dispatching equivalent model construction method in any of the above-mentioned implementation modes.

[0071] The second determination module 43 is used to determine the target virtual power plant based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis result of the dispatching instruction, and adopt the target virtual power plant to execute the active power dispatching instruction.

[0072] The active power dispatching device for a virtual power plant provided in an embodiment of the present application can, when the virtual power plant's power generation plan has been formulated a day ago, determine the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active power real-time dispatching instruction of the power grid through the constructed active power real-time dispatching cost model and active power real-time dispatching instruction feasibility model. Therefore, for a given active power grid real-time dispatching instruction, the active power real-time dispatching cost of each virtual power plant can be quickly determined, thereby facilitating the active power real-time dispatching of the virtual power plant by the power grid.

[0073] According to an embodiment of the present application, the present application also provides an electronic device.

[0074] Figure 5A schematic block diagram of an example electronic device that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0075] like Figure 5 As shown, the electronic device includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. RAM 803 can also store various programs and data required for the operation of the electronic device. The computing unit 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to bus 804.

[0076] Multiple components in the electronic device are connected to the I / O interface 805, including an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows the electronic device to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0077] The computing unit 801 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the method for constructing a value model for real-time active power dispatch of a virtual power plant or the method for dispatching active power of a virtual power plant. For example, in some embodiments, the method for constructing a value model for real-time active power dispatch of a virtual power plant or the method for dispatching active power of a virtual power plant can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device via the ROM 802 and / or the communication unit 809. When the computer program is loaded into RAM 803 and executed by computing unit 801, one or more steps of the method for constructing a value model for real-time active power dispatch of a virtual power plant or the method for dispatching active power of a virtual power plant described above can be executed. Alternatively, in other embodiments, computing unit 801 can be configured to execute the method for constructing a value model for real-time active power dispatch of a virtual power plant or the method for dispatching active power of a virtual power plant using any other appropriate means (e.g., via firmware).

[0078] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0079] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0080] The present invention provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to execute the method for constructing an equivalent model for active power scheduling of a virtual power plant or the method for active power scheduling of a virtual power plant described in the present invention.

[0081] An embodiment of the present application provides a computer-readable storage medium storing executable instructions, wherein the executable instructions are stored. When the executable instructions are executed by a processor, the processor will execute the method for constructing an equivalent model of active power scheduling of a virtual power plant, or the active power scheduling method of a virtual power plant, provided in an embodiment of the present application.

[0082] In some embodiments, a computer-readable storage medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium may be a machine-readable signal medium or a machine-readable storage medium. The computer-readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0083] In some embodiments, executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0084] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinating files (e.g., files storing one or more modules, subroutines, or code portions).

[0085] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0086] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0087] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0088] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0089] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.

[0090] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0091] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for constructing a virtual power plant active real-time dispatch equivalent model, characterized in that: include: Acquire multiple real-time training active scheduling instructions, wherein the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power; Obtaining a real-time optimization scheduling model for active power within the virtual power plant; the real-time optimization scheduling model for active power within the virtual power plant satisfies the requirements of training real-time active power scheduling instructions and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost of the virtual power plant within a scheduling cycle; Based on the real-time active power optimization scheduling model within the virtual power plant and the plurality of training real-time active power scheduling instructions, determining corresponding multiple training real-time scheduling costs and multiple training scheduling instruction feasibility analysis results; Based on the plurality of the training active real-time scheduling instructions and the corresponding plurality of the training real-time scheduling costs, training the first neural network to obtain an active real-time scheduling cost model; Based on the plurality of training active real-time scheduling instructions and the corresponding plurality of training scheduling instruction feasibility analysis results, a second neural network is trained to obtain an active real-time scheduling instruction feasibility model; Based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model, the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instruction of the power grid are determined.

2. The method for constructing an equivalent model of active real-time scheduling of a virtual power plant according to claim 1, characterized in that: Obtain the real-time optimization scheduling model of active power within the virtual power plant, including: Obtaining an objective function corresponding to the real-time optimization scheduling model for active power within the virtual power plant; the objective function is related to a first operating cost function of the thermal power source within the virtual power plant within a scheduling cycle and a second operating cost function of the energy storage system power source within a scheduling cycle; Determine the constraints corresponding to the objective function; the constraints include a first operating constraint corresponding to the thermal power source, a second operating constraint corresponding to the energy storage system power source, and a power balance constraint within the virtual power plant.

3. The method for constructing an equivalent model of active real-time scheduling of a virtual power plant according to claim 2, characterized in that: The internal power balance constraints of the virtual power plant include: During the training scheduling period, the sum of the output power of the thermal power supply inside the virtual power plant, the output power of the energy storage system power supply, and the training scheduling power corresponding to the training active real-time scheduling instructions is equal to the total power of the internal load of the virtual power plant.

4. The method for constructing an equivalent model of active real-time scheduling of a virtual power plant according to claim 1, characterized in that: Multiple real-time scheduling instructions for training were obtained, including: Determine the scheduling start time range, scheduling duration range, and scheduling power range corresponding to the training active real-time scheduling instruction; Based on the scheduling start time range, the scheduling duration time range and the scheduling power range, a plurality of the training active real-time scheduling instructions are randomly generated using a uniform distribution function.

5. The method for constructing an equivalent model of active real-time scheduling of a virtual power plant according to claim 1, characterized in that: Based on the real-time active power optimization scheduling model within the virtual power plant and the plurality of training real-time active power scheduling instructions, determining corresponding plurality of training real-time scheduling costs and plurality of training scheduling instruction feasibility analysis results, including: Based on each of the trained active real-time scheduling instructions, the active real-time optimization scheduling model within the virtual power plant is solved to obtain a solution result; the solution result includes a first output power curve of the thermal power source within the virtual power plant during the training scheduling duration, a second output power curve of the energy storage system power source during the training scheduling duration, and a minimum real-time scheduling cost of the virtual power plant training within a corresponding scheduling cycle; If the solution result shows that the first output power curve, the second output power curve, and the training minimum real-time scheduling cost have a solution, the training minimum real-time scheduling cost is used as the corresponding training real-time scheduling cost; and the feasibility analysis result of the training active scheduling instruction is determined to be feasible; If the solution result indicates that there is no solution for the first output power curve, the second output power curve or the training minimum real-time scheduling cost, the feasibility analysis result of the training active scheduling instruction is determined to be infeasible.

6. A virtual power plant active power scheduling method, characterized in that: include: Obtaining active power dispatch instructions from the power grid to the virtual power plant, wherein the active power dispatch instructions include a dispatch start period, a dispatch duration period, and a dispatch power; Based on the active real-time dispatch cost model and active real-time dispatch instruction feasibility model corresponding to each virtual power plant, the active dispatch instructions are processed respectively to obtain the real-time dispatch cost and dispatch instruction feasibility analysis results when each virtual power plant executes the active dispatch instruction; the active real-time dispatch cost model and active real-time dispatch instruction feasibility model are constructed using the virtual power plant active real-time dispatch equivalent model construction method according to any one of claims 1 to 5; Based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis results of the dispatching instruction, a target virtual power plant is determined, and the target virtual power plant is used to execute the active power dispatching instruction.

7. The method for active power dispatching of a virtual power plant according to claim 6, characterized in that: Based on the real-time dispatch costs and dispatch instruction feasibility analysis results of each virtual power plant, the target virtual power plant is determined, including: Based on the real-time dispatching costs and dispatching instruction feasibility analysis results corresponding to each virtual power plant, the virtual power plant with the feasible dispatching instruction feasibility analysis result and the lowest real-time dispatching cost is determined as the target virtual power plant.

8. A device for constructing an equivalent model for real-time active power dispatch of a virtual power plant, characterized in that: include: A first acquisition module is configured to acquire a plurality of real-time training active scheduling instructions, wherein the real-time training active scheduling instructions include a training scheduling start period, a training scheduling duration period, and a training scheduling power; The second acquisition module is used to obtain the real-time optimization scheduling model of the active power within the virtual power plant; the real-time optimization scheduling model of the active power within the virtual power plant satisfies the requirements of training the real-time active power scheduling instructions and determines the planned output power curves of various types of distributed power sources within the virtual power plant within a scheduling cycle, with the goal of minimizing the real-time scheduling cost of the virtual power plant within a scheduling cycle; A first determination module is configured to determine a plurality of corresponding training real-time scheduling costs and a plurality of feasibility analysis results of the training scheduling instructions based on the real-time active power optimization scheduling model within the virtual power plant and the plurality of training real-time active power scheduling instructions; A first training module is configured to train a first neural network based on the plurality of training active real-time scheduling instructions and the corresponding plurality of training real-time scheduling costs to obtain an active real-time scheduling cost model; The second training module is used to train the second neural network based on multiple trained active real-time dispatching instructions and corresponding multiple training dispatching instruction feasibility analysis results to obtain an active real-time dispatching instruction feasibility model; and to determine the real-time dispatching cost and dispatching instruction feasibility analysis results when the virtual power plant executes the active real-time dispatching instructions of the power grid based on the active real-time dispatching cost model and the active real-time dispatching instruction feasibility model.

9. A virtual power plant active power dispatching device, characterized in that: include: The third acquisition module is used to obtain the active power dispatch instruction of the power grid to the virtual power plant, wherein the active power dispatch instruction includes the dispatch start period, the dispatch duration period and the dispatch power; a processing module for processing the active power dispatch instructions respectively based on the active power real-time dispatch cost model and the active power real-time dispatch instruction feasibility model corresponding to each virtual power plant, to obtain the real-time dispatch cost and dispatch instruction feasibility analysis results when each virtual power plant executes the active power dispatch instruction; the active power real-time dispatch cost model and the active power real-time dispatch instruction feasibility model are constructed using the virtual power plant active power real-time dispatch equivalent model construction method according to any one of claims 1 to 5; The second determination module is used to determine the target virtual power plant based on the real-time dispatching cost corresponding to each virtual power plant and the feasibility analysis result of the dispatching instruction, and adopt the target virtual power plant to execute the active power dispatching instruction.

10. An electronic device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the method for constructing an equivalent model of real-time active power scheduling of a virtual power plant as described in any one of claims 1 to 5, or the method for active power scheduling of a virtual power plant as described in claim 6 or 7.

Images