Virtual power plant optimization scheduling method, electronic equipment and storage medium

Abstract

The invention discloses a virtual power plant optimal scheduling method, electronic equipment and a storage medium. The method comprises the following steps: acquiring associated data corresponding to distributed energy and a gas turbine power generation system in a virtual power plant; wherein the distributed energy source comprises a distributed power supply, an energy storage system and a user flexible load; establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; for user flexible loads in the virtual power plant, determining a target user load combination participating in optimal scheduling of the virtual power plant through a dobby machine algorithm; and determining an optimal scheduling scheme of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination. According to the technical scheme, the optimal target user load combination is selected by adopting the dobby machine algorithm, so that the aggregation accuracy of the virtual power plant is improved, and the range of the affected users is reduced. The overall operation benefit of the virtual power plant can be improved, and powerful support can be provided for stable operation of a power system.

Description

Technical Field

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

[0002] The challenges of energy security and climate change facing the world are becoming increasingly severe. In order to cope with these challenges, distributed renewable energy has become the core development direction of energy in many countries. In order to accelerate the optimal combination of coal and new energy, countries are actively promoting the development of distributed energy resources (DERs) and are committed to building a new power system.

[0003] As an advanced power system management concept, Virtual Power Plant (VPP) can provide higher flexibility, reliability and efficiency by integrating and coordinating distributed resources, so as to meet the needs of the power system more flexibly. The virtual power plant with coordinated main and micro resources can realize the aggregation and coordinated optimization of distributed energy sources such as distributed power sources, energy storage systems, and controllable flexible loads. Among them, controllable flexible loads have the characteristics of wide distribution, fast frequency regulation response speed, large potential frequency regulation capacity, continuous and regular working time, and small impact of short-term control on users. If appropriate control strategies can be adopted to aggregate such massive and dispersed demand-side loads, it can achieve the effect of improving energy utilization efficiency and providing auxiliary service backup. Therefore, the present invention focuses on the participation of load-type virtual power plants in scheduling, and releases the flexible regulation potential of load resources by effectively aggregating dispersed load resources.

[0004] Domestic and foreign scholars have conducted a series of studies on VPP participation in the electricity market, mainly focusing on bidding strategies and game models. However, most of the current research ignores the uncertainty of the user's flexible load in the virtual power plant in response. This uncertainty is caused by the environment or other external factors, which may lead to incomplete load response or even exit the response mechanism, thereby causing deviations in the aggregated power of the virtual power plant, thereby affecting the stability of the system and the economic benefits of the VPP. Summary of the invention

[0005] The present invention provides a virtual power plant optimization scheduling method, electronic equipment and storage medium, so as to improve the scheduling efficiency and economy of the virtual power plant and optimize the configuration and use of energy resources.

[0006] According to one aspect of the present invention, a virtual power plant optimization scheduling method is provided, the method comprising:

[0007] Acquire the associated data corresponding to the distributed energy and the gas turbine power generation system in the virtual power plant; wherein the distributed energy includes distributed power sources, energy storage systems and user flexible loads;

[0008] Establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; wherein the virtual power plant mathematical model includes an operation cost model and an operation income model of the virtual power plant;

[0009] For the flexible loads of users in the virtual power plant, a target user load combination participating in the optimal scheduling of the virtual power plant is determined by a multi-armed bandit algorithm;

[0010] An optimal scheduling scheme for the virtual power plant is determined based on the associated data, the mathematical model of the virtual power plant and the target user load combination.

[0011] According to another aspect of the present invention, a virtual power plant optimization scheduling device is provided, the device comprising:

[0012] A correlation data acquisition module is used to acquire correlation data corresponding to distributed energy and gas turbine power generation systems in a virtual power plant; wherein the distributed energy includes distributed power sources, energy storage systems and user flexible loads;

[0013] A power plant model building module, used to build a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; wherein the virtual power plant mathematical model includes an operation cost model and an operation income model of the virtual power plant;

[0014] A user load determination module is used to determine a target user load combination participating in the optimal scheduling of the virtual power plant through a multi-armed bandit algorithm for the user flexible load in the virtual power plant;

[0015] The power plant optimization scheduling module is used to determine the optimization scheduling plan of the virtual power plant based on the associated data, the virtual power plant mathematical model and the target user load combination.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] at least one processor;

[0018] and a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the virtual power plant optimization scheduling method described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the virtual power plant optimization scheduling method described in any embodiment of the present invention when executed.

[0021] The technical solution of the embodiment of the present invention includes obtaining the associated data corresponding to the distributed energy and gas turbine power generation system in the virtual power plant; establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; determining the target user load combination participating in the optimal scheduling of the virtual power plant through a multi-armed slot machine algorithm for the user flexible load in the virtual power plant; and determining the optimal scheduling scheme of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination. In the embodiment of the present invention, in order to solve the technical problem that the uncertainty of the user flexible load response behavior causes the aggregated power of the virtual power plant to deviate from the expected aggregation, affecting the stability of the power system and the economic benefits of the virtual power plant, a multi-armed slot machine algorithm is used to select the optimal target user load combination, thereby improving the aggregation accuracy of the virtual power plant and reducing the scope of affected users. It can not only improve the overall operating efficiency of the virtual power plant, but also provide strong support for the stable operation of the power system.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0024] Figure 1 A flowchart of a virtual power plant optimization scheduling method provided in Embodiment 1 of the present invention;

[0025] Figure 2 A flowchart of a virtual power plant optimization scheduling method provided in Embodiment 2 of the present invention;

[0026] Figure 3 A graph showing the relationship between the number of learning times and the load exit rate provided in the second embodiment of the present invention;

[0027] Figure 4 This is a diagram of optimized power distribution of a 24h virtual power plant provided in the second embodiment of the present invention;

[0028] Figure 5A comparison diagram of the power load before and after optimization provided in the second embodiment of the present invention;

[0029] Figure 6 A schematic diagram of the structure of a virtual power plant optimization scheduling device provided in Embodiment 3 of the present invention;

[0030] Figure 7 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 described embodiments 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 creative work should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] Embodiment 1

[0034] Figure 1 This is a flowchart of a virtual power plant optimization scheduling method provided in the first embodiment of the present invention. This embodiment can be applied to optimize the scheduling of distributed energy in a virtual power plant to improve the scheduling efficiency and economy of the virtual power plant and optimize the configuration and use of resources. The method can be executed by a virtual power plant optimization scheduling device, which can be implemented in the form of hardware and / or software and can be configured in a computer device. Figure 1 As shown, the method comprises the following steps:

[0035] S110. Obtaining associated data corresponding to distributed energy and gas turbine power generation systems in the virtual power plant.

[0036] Among them, the virtual power plant can be a virtual power plant with main distribution and micro-cooperation, which combines the collaborative interaction between the main grid, distribution network and microgrid, and realizes efficient power regulation and energy utilization through the technical means of the virtual power plant. Distributed energy can include distributed power sources, energy storage systems and user flexible loads. Distributed power sources include distributed photovoltaic and wind power. Associated data can be data related to distributed energy and gas turbine power generation systems in virtual power plants. Associated data include but are not limited to the upper and lower limits of the output of each unit in the virtual power plant, the total system load, and time-of-use electricity prices.

[0037] Specifically, data collection equipment, such as smart meters and sensors, can be deployed at the distributed energy and gas turbine power generation system to collect relevant data in real time. Then, based on communication technology, the collected relevant data is transmitted to the data storage system of the virtual power plant, and the received relevant data is stored and managed. The relevant data can provide a data basis for the establishment of the virtual power plant mathematical model and the scheduling of the virtual power plant.

[0038] S120. Establish a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data.

[0039] The mathematical model of the virtual power plant may include an operating cost model and an operating income model of the virtual power plant. The operating cost model refers to a mathematical model that represents the daily operating cost of the virtual power plant, and the operating income model refers to a mathematical model that represents the daily operating income of the virtual power plant. In order to improve the economic benefits of the virtual power plant, a mathematical model of the virtual power plant including distributed power sources, energy storage systems, user flexible loads, and gas turbine power generation systems can be established based on associated data.

[0040] In an embodiment of the present invention, a mathematical model of a virtual power plant corresponding to the virtual power plant is established based on associated data, which may include: establishing an operating cost model of the virtual power plant based on the operating and depreciation costs of distributed power sources, the cost of purchasing electricity from the power grid, the cost of using fuel for gas turbines, and compensation costs for optimizing scheduling of user flexible loads.

[0041] It should also be noted that the compensation for optimizing the scheduling of users' flexible loads can be understood as a kind of economic compensation provided by power companies to users in order to encourage users to actively participate in the operation and control of the power grid and adjust their own electricity consumption to cooperate with the supply and demand balance and optimized scheduling of the power grid.

[0042] Specifically, users can reach an agreement with the power company to reduce or increase their own electricity consumption during peak or off-peak periods or other agreed periods, thereby facilitating the power company's load management. The power company pays compensation fees to the users who have agreed to reduce electricity consumption, or reduces electricity prices to encourage electricity consumption, etc. This part of the fee can be listed as compensation for optimizing the dispatch of users' flexible loads.

[0043] For example, the daily operating costs of a virtual power plant with main power distribution and micro-cooperation are mainly composed of the following parts: the operation and depreciation costs of distributed power sources, the cost of the virtual power plant purchasing electricity from the power grid, the cost of fuel used by gas turbines, and the compensation for optimizing the dispatch of user flexible loads. The above parts constitute the sum of the operating costs of the virtual power plant, so the operating cost model of the virtual power plant can be expressed by the following formulas (1)-(5):

[0044] C VPP =C DG +C Buy +C MT +C L (1)

[0045]

[0046] C L =K Cut P AL (t) (5)

[0047] Among them, C VPP represents the total cost of the virtual power plant; C DG is the operating cost of distributed power sources, including the operation and maintenance costs of distributed photovoltaic and wind power; C Buy C is the cost of purchasing electricity from the grid for the virtual power plant; MT is the power generation cost of the gas turbine; C L Compensation for optimizing the scheduling of user flexible loads.

[0048] T is the time set; K PV and K W are the operating cost coefficients of distributed photovoltaic and wind power respectively, K B is the depreciation coefficient of energy storage; P PV (t) and P W (t) is the output power of distributed photovoltaic and wind turbines, P B (t) is the charging and discharging power of the energy storage, charging is positive and discharging is negative.

[0049] K Buy is the time-of-use electricity price of the virtual power plant purchasing electricity from the grid, P Buy (t) is the power purchased from the power grid; K MT is the operating cost coefficient of the gas turbine, P MT (t) is the output power of the gas turbine; K Cut is the cost coefficient of user flexible load dispatch, that is, the contract unit price of load participating in demand response, P AL (t) is the actual response of the user's flexible load.

[0050] It is understandable that the actual response of user flexible load refers to the load change caused by the user actually changing his electricity consumption behavior according to the incentive mechanism or grid dispatch instructions. For example, during peak hours, users can choose to reduce electricity consumption to reduce the power supply pressure of the grid; during low electricity consumption hours, users can increase electricity consumption to make full use of the remaining capacity of the grid.

[0051] In an embodiment of the present invention, a mathematical model of a virtual power plant corresponding to the virtual power plant is established based on associated data, which may include: establishing an operating income model of the virtual power plant based on the operating income of the distributed power source corresponding to the virtual power plant, the income from selling electricity to the power grid, the power generation income of the gas turbine, and the penalty for no response or incomplete response of the user's flexible load.

[0052] In the practical application of the present invention, distributed power sources include but are not limited to distributed photovoltaics, wind power, etc. The operating income of distributed power sources refers to the direct operating income that distributed power sources bring to virtual power plants through grid-connected power generation. Furthermore, virtual power plants can also sell the power generated by distributed power sources to the power grid, thereby obtaining electricity sales income. In some virtual power plants, gas turbines are also one of the important power generation resources, and the power generation income of gas turbines is also an important part of the daily operating income of virtual power plants.

[0053] In the operation of virtual power plants, in order to ensure the stable operation of the power grid and the balance of supply and demand, it is necessary to accurately control the flexible loads of users participating in the dispatch. However, in actual operation, users may not be able to fully respond to dispatch instructions or even exit the response mechanism due to various reasons, such as equipment failure and production plan adjustment, which means there is uncertainty. At this time, the virtual power plant can impose certain punitive charges on users based on the relevant agreements signed with users, and this part of the charges can also be used as the operating income cost of the virtual power plant.

[0054] For example, the daily operating income of a virtual power plant with main power distribution and micro-cooperation is mainly composed of the following parts: operating income of distributed power sources, income from selling electricity to the grid, power generation income of gas turbines, and penalties for users not responding or responding incompletely in actual scheduling. The above parts constitute the sum of the operating income of the virtual power plant, so the operating cost model of the virtual power plant can be expressed by the following formulas (6)-(10):

[0055] R VPP =R DG +R Sell +R MT +R L (6)

[0056]

[0057] RL =J Cut [P L (t)-P AL (t)] (10)

[0058] Among them, R VPP represents the total revenue of the virtual power plant; R DG is the operating income of distributed power sources, including the power generation income of distributed photovoltaic and wind power; R Sell is the revenue from selling electricity from the virtual power plant to the grid; R MT is the power generation income of the gas turbine; R L Compensation cost for flexible loads.

[0059] J PV and J W are the operating cost coefficients of distributed photovoltaic and wind power respectively; J Sell is the time-of-use electricity price for selling electricity to the power grid; J MT is the power generation profit coefficient of the gas turbine; J B is the charging and discharging income coefficient of energy storage; J Cut is the penalty factor of the virtual power plant on the user; P Sell (t) is the electricity sold; P L is the target response power of the aggregated user.

[0060] It should also be noted that the target response power of aggregated users can refer to the expected power adjustment target set by the virtual power plant for the aggregated user group based on factors such as the supply and demand conditions of the power grid and renewable energy generation forecasts. It can be used to guide users to adjust their electricity consumption behavior to achieve supply and demand balance and optimized scheduling of the power grid.

[0061] Combining the above content, the profit of the virtual power plant is shown in formula (11):

[0062] P VPP =R VPP -C VPP (11)

[0063] S130. For the flexible loads of users in the virtual power plant, a target user load combination participating in the optimal scheduling of the virtual power plant is determined by a multi-armed bandit algorithm.

[0064] It is understandable that user flexible loads have the characteristics of wide distribution, fast frequency regulation response speed, large potential frequency regulation capacity, continuous and regular working hours, and small impact of short-term control on users. If appropriate control strategies can be adopted to aggregate such massive and dispersed loads, energy utilization efficiency can be improved. However, in the actual scheduling of virtual power plants, there is uncertainty in the response of user flexible loads to the scheduling of virtual power plants. This uncertainty is caused by the environment or other external factors, which may lead to incomplete response of the load or even exit the response mechanism, thereby causing deviations in the aggregated power of the virtual power plant, which in turn affects the stability of the system and the economic benefits of the virtual power plant.

[0065] In order to solve this problem, the uncertainty of user response behavior is taken into account in the embodiment of the present invention, and a multi-armed tiger machine algorithm is used to select users, and the optimal user combination is selected from a large number of user flexible loads as the target user load combination participating in the optimal scheduling of the virtual power plant. The target user load combination determined through this process has a higher responsiveness in the scheduling optimization process of the virtual power plant, and there are fewer cases of incomplete response or exit from the response mechanism, thereby improving the accuracy of aggregation and reducing the scope of affected users.

[0066] S140. Determine an optimal dispatching plan for the virtual power plant based on the associated data, the mathematical model of the virtual power plant and the target user load combination.

[0067] It can be understood that the flexible load and actual response power of users participating in the scheduling can be determined based on the target user load combination, and then the optimal scheduling plan can be determined with the goal of maximizing profits by combining the related data and the mathematical model of the virtual power plant. This can not only significantly improve the scheduling efficiency and economy of the virtual power plant, but also optimize the configuration and use of energy resources.

[0068] In an embodiment of the present invention, an optimal scheduling scheme for a virtual power plant is determined based on associated data, a mathematical model of a virtual power plant, and a target user load combination, including: determining the optimal scheduling constraints corresponding to the virtual power plant based on the associated data; determining a target cost function corresponding to the virtual power plant based on the mathematical model of the virtual power plant; and determining the optimal scheduling scheme for the virtual power plant based on the target cost function, the target user load combination, and the optimal scheduling constraints.

[0069] Among them, the optimization scheduling constraints include distributed generation constraints, power exchange constraints with the power grid, gas turbine power generation constraints, load power upper and lower limit constraints and energy storage constraints.

[0070] For example, the optimal dispatching scheme of the virtual power plant can be obtained by optimizing the solution with the goal of maximizing the total operating profit. The power supply and demand situation for the next day or a period of time in the future can be predicted and analyzed, and the optimal dispatching strategy can be formulated to achieve the internal supply and demand balance and maximize the profit of the virtual power plant.

[0071] The objective cost function is to maximize the profit of the virtual power plant, as shown in formula (12):

[0072] min(R VPP -C VPP )(12)

[0073] When scheduling, the virtual power plant with coordinated power distribution and micro-power generation must consider the power balance constraint to ensure the balance of the total output power of all distributed resources under its unified management, the power of the power generation system, the load it bears, and the power bought and sold in the transaction, that is:

[0074] P W +P PV +P Buy +P MT =P Sell +P B +P L -P AL (13)

[0075] Among them, the battery power is P B =P C -P D In the formula, P C Indicates the charging power of energy storage; P D is the discharge power of the energy storage.

[0076] The multi-period scheduling of the virtual power plant with coordinated main and micro power generation must meet the requirements of distributed generation (distributed photovoltaic (14), wind power generation constraints (15), energy storage charging and discharging power constraints (16)-(17)), power exchange constraints with the power grid (18)-(19), gas turbine power generation constraints (20), and load power upper and lower limit constraints (21).

[0077]

[0078]

[0079] The state of charge (SOC) of the energy storage meets the upper and lower limit constraints:

[0080]

[0081] In the formula, S SOC and are the minimum and maximum values ​​of the energy storage charge state, respectively.

[0082] The capacity of energy storage must meet upper and lower limit constraints:

[0083]

[0084] In the formula, E SOC and are the minimum and maximum values ​​of the energy storage capacity respectively.

[0085] Since the storage energy cannot be in both charging and discharging states at the same time, the following constraints are met:

[0086] X C (t)·X D (t)=0 (24)

[0087] Where, X C and X D are charging state and discharging state respectively, and X C ∈{0,1},X D ∈{0,1}.

[0088] Since the number of energy storage charge and discharge times will affect the lifespan, constraints on the number of charge and discharge times are added:

[0089]

[0090] Where N C and N D They are respectively the charging and discharging times limits of energy storage.

[0091] On the basis of the above technical scheme, based on the target cost function, the target user load combination and the optimized scheduling constraints, the optimal scheduling scheme of the virtual power plant is determined, which may include: according to the target cost function, the target user load combination and the optimized scheduling constraints, determining the scheduling parameters corresponding to the distributed energy and gas turbine power generation system in the virtual power plant when the value of the target cost function is minimized through a commercial solver.

[0092] Specifically, the solution can be based on commercial solvers, such as GUROBI (Gurobi Optimizer), CP LEX (digital optimization technology), etc., to obtain the optimal scheduling plan for the virtual power plant. The target cost function, target user load combination and optimization scheduling constraints can be input into the commercial solver. The solver will use advanced algorithms and numerical techniques to find a scheduling plan that meets all constraints and minimizes the target cost function value. During the solution process, the solver will consider various possible scheduling parameter combinations and evaluate their impact on the target cost function. When the solver finds the optimal solution, it will output a set of scheduling parameters, including but not limited to the power output of distributed energy and gas turbine power generation systems, the charging and discharging status of energy storage devices, etc. These scheduling parameters constitute the optimal scheduling plan for the virtual power plant, which can be used to guide operation and scheduling in actual operation.

[0093] The technical solution of the embodiment of the present invention includes obtaining the associated data corresponding to the distributed energy and gas turbine power generation system in the virtual power plant; establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; determining the target user load combination participating in the optimal scheduling of the virtual power plant through a multi-armed slot machine algorithm for the user flexible load in the virtual power plant; and determining the optimal scheduling scheme of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination. In the embodiment of the present invention, in order to solve the technical problem that the uncertainty of the user flexible load response behavior causes the aggregated power of the virtual power plant to deviate from the expected aggregation, affecting the stability of the power system and the economic benefits of the virtual power plant, a multi-armed slot machine algorithm is used to select the optimal target user load combination, thereby improving the aggregation accuracy of the virtual power plant and reducing the scope of affected users. It can not only improve the overall operating efficiency of the virtual power plant, but also provide strong support for the stable operation of the power system.

[0094] Embodiment 2

[0095] Figure 2 This is a flowchart of a virtual power plant optimization scheduling method provided in the second embodiment of the present invention. Based on the above embodiment, this embodiment further refines the process of determining the target user load combination. The specific implementation method can refer to the technical solution of this embodiment. Among them, the technical terms that are the same or corresponding to the above embodiment are not repeated here.

[0096] In the embodiment of the present invention, the uncertainty of the user's flexible load response is considered, and the load optimization problem is modeled as a combinatorial multi-armed bandit problem. The virtual power plant aggregation problem of the main distribution micro-cooperation is modeled as a combinatorial multi-armed bandit (CMAB) model, and the goal is to minimize the regret value, that is, the difference between the actual aggregation amount and the best possible strategy aggregation amount. The best possible strategy is the offline optimal algorithm (OfflineOptimization).

[0097] In an embodiment of the present invention, it is assumed that each user has a probability of responding or choosing to exit, and they are independent of each other. These independent decisions will directly affect the aggregation effect of the virtual power plant, and the user load response is modeled as a probability function that follows the Bernoulli distribution.

[0098] like Figure 2 As shown, the method comprises the following steps:

[0099] S210. Obtaining associated data corresponding to distributed energy and gas turbine power generation systems in the virtual power plant.

[0100] Among them, distributed energy includes distributed power sources, energy storage systems and user flexible loads.

[0101] S220. Establish a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data.

[0102] Among them, the mathematical model of the virtual power plant includes the operating cost model and operating income model of the virtual power plant.

[0103] S230. For the user flexible load in the virtual power plant, determine the benchmark user load combination participating in the optimal scheduling of the virtual power plant based on the offline optimal algorithm, and calculate the first loss value corresponding to the benchmark user load combination.

[0104] The reference user load combination may be a user load combination determined by an offline optimal algorithm. The first loss value may represent a difference between an actual response power and an expected power of the reference user load combination.

[0105] In an embodiment of the present invention, a benchmark user load combination participating in the optimal scheduling of a virtual power plant is determined based on an offline optimal algorithm, and a first loss value corresponding to the benchmark user load combination is calculated, including: determining a preset response probability of each user flexible load in the virtual power plant based on a probability distribution function; determining at least two benchmark user loads from a plurality of user flexible loads based on the preset response probability and the offline optimal algorithm; combining at least two benchmark user loads into a benchmark user load combination; and calculating a first loss value based on the preset response probability and power value corresponding to each benchmark user load and a target aggregate power value.

[0106] It can be understood that the user load response is modeled as a probability function that follows the Bernoulli distribution, that is, a probability distribution function. The response probability of each user's flexible load in the virtual power plant, that is, the preset response probability, can be determined based on the probability distribution function. When the response probability of each user's flexible load is known, the offline algorithm determines at least two benchmark user loads from multiple user flexible loads to form a benchmark user load combination based on the most ideal situation, that is, selects the optimal subset. Further, the first loss value is calculated according to the preset response probability and power value of each benchmark user load in the benchmark user load combination and the target aggregate power value. Among them, the target aggregate power value can be the power value expected to be aggregated in the virtual power plant scheduling. Specifically, the preset response probability and power value of the benchmark user load can be multiplied to obtain the aggregate power value of the benchmark user load combination. Then, the aggregate power value and the target aggregate power value are substituted into the first loss function to obtain the first loss value. The first loss function can be a cross entropy loss function.

[0107] S240. Determine a candidate user load combination participating in the optimal scheduling of the virtual power plant based on the CUCB-Avg algorithm, and calculate a second loss value corresponding to the candidate user load combination.

[0108] In this embodiment, the CUCB-Avg algorithm is an online learning algorithm that combines an upper confidence bound (UpperConfidence Bound) and a sample average (Average) to balance the trade-off between exploration and exploitation. The CUCB-Avg algorithm is used to solve the combined multi-armed bandit problem. The candidate user load combination can be a user load combination determined by the CUCB-Avg algorithm, and the second loss value can be used to represent the difference between the actual aggregate power of the candidate user load combination and the target aggregate power value.

[0109] is the loss value of algorithm z at time t, that is, the second loss value, as shown in formula (27):

[0110]

[0111] Among them, algorithm z is the CUCB-Avg algorithm, P i is the power value of user i; X t,i is the response of user i at time t; is the user set selected by algorithm z at time t, i.e., the candidate user load combination; D t is the target aggregate power value at time t.

[0112] S250. Determine a target regret value based on the first loss value and the second loss value, and when the target regret value meets a preset condition, use the candidate user load combination as the target user load combination.

[0113] Specifically, the results obtained by the offline optimal algorithm can be used as a benchmark, and the performance difference between the offline optimal algorithm and the online learning algorithm CUCB-Avg over a period of time can be compared, that is, the target regret value, which is defined as shown in formula (28):

[0114]

[0115] in, is the loss of the offline optimal algorithm at time t, that is, the first loss value; T is the total time step; is the loss of algorithm z at time t, that is, the second loss value.

[0116] It should also be noted that the offline optimal algorithm, since the response probability value of each user's flexible load is known, is equivalent to obtaining the optimal combination under ideal conditions, that is, the benchmark user load combination. Furthermore, the candidate user load combination is also determined through an online learning algorithm, such as the CUCB-Avg algorithm. Finally, the target regret value is calculated based on the loss values ​​of the two algorithms. If the target regret value meets the preset conditions, for example, the target regret value is less than the preset threshold, it means that the candidate user load combination determined by the CUCB-Avg algorithm is better and can be used as the target user load combination, that is, as the combination of user flexible loads that ultimately participate in the virtual power plant scheduling optimization.

[0117] In an embodiment of the present invention, when the target regret value does not meet the preset conditions, the following process can be executed in a loop: updating the response probability and the number of selections of each candidate user load in the candidate user load combination; determining the current user load combination based on the updated response probability and the number of selections, and calculating the current loss value corresponding to the current user load combination; updating the second loss value based on the current loss value, and determining the target regret value based on the first loss value and the updated second loss.

[0118] Specifically, when the target regret value does not meet the preset conditions, the response probability and the number of selections can be updated. The number of selections refers to the number of times the user flexible load is selected when determining the candidate user load combination. Then, the current user load combination is determined based on the updated response probability and the number of selections, and the current loss value corresponding to the current user load combination is calculated.

[0119] The current user load combination refers to the new candidate user load combination re-determined after the response probability and the number of selections are updated. The current loss value refers to the loss value corresponding to the current user load combination, and the current loss value can be used as the new second loss value. Then, the target regret value is determined based on the first loss value and the updated second loss. The above process can be repeated until the current user load combination corresponding to the target regret value pair that meets the preset conditions is determined as the target user load combination.

[0120] In practical applications, the target user load combination and the actual aggregate power are learned based on the CUCB-Avg algorithm, including: the CUCB-Avg algorithm combines the sample average and the confidence upper limit to make decisions. The algorithm consists of two parts: 1) sorting the users in non-increasing order according to the confidence upper limit; 2) selecting the least users according to the sorting and the target value at this time. The CUCB-Avg algorithm process is shown in Table 1.

[0121] Table 1

[0122]

[0123]

[0124] The CMAB problem is to balance exploration and preservation. i (t) consists of two parts, the first one represents conservation and the second one represents exploration. α>0 is a parameter used to balance exploration and conservation in the CMAB problem; T i (t) is the number of times the i-th user is selected at time t; where the algorithm estimates the probability of each user responding p i The calculation of (t) is shown in formula (29):

[0125]

[0126] In the formula, I i (t) is the set of users selected at time t.

[0127] S260. Determine an optimal dispatching plan for the virtual power plant based on the associated data, the mathematical model of the virtual power plant and the target user load combination.

[0128] In the embodiment of the present invention, during the CUCB-Avg learning process, as the number of learning times increases, the exit rate of the selected users gradually shows a downward trend, such as Figure 3 As shown, it is a relationship diagram between the number of learning times and the load exit rate provided in the second embodiment of the present invention, indicating that as the learning goes deeper, the understanding of the user is continuously enhanced, thereby reducing the number of user exit responses.

[0129] like Figure 4 As shown, it is a diagram of the optimized power distribution of the 24h virtual power plant provided by an embodiment of the present invention. During the night valley period (00:00-07:00), due to the low load, the electricity price is also at a low point, the price is 0.25 yuan per kilowatt-hour. During this period, due to the lack of photovoltaic output, wind power and gas turbines become the main suppliers of electricity. The gas turbine adopts a heat-to-electricity operation mode, which not only maintains a stable power output, but also has excess power to charge the battery. The power load increases significantly during the daytime peak period (10:00-15:00 and 18:00-21:00), but the distributed photovoltaic energy output is sufficient during the day. Despite the surge in electricity demand, the virtual power plant still has excess power to sell to the grid. In addition, during these peak periods, the batteries that were previously sufficient during the valley period begin to release electricity to provide the necessary power support for the virtual power plant. The energy storage system effectively utilizes the difference between peak and valley electricity prices by charging when the electricity price is low and releasing electricity when the electricity price is peak.

[0130] like Figure 5As shown, it is a comparison diagram of the power load before and after optimization provided in the second embodiment of the present invention. It can be clearly seen that the load response during the daytime (11:00-15:00) and nighttime (19:00-21:00) periods is the most significant. There are more load responses in the period from 11:00 to 15:00, mainly because commercial activities are reduced at this time, and the power demand of most residents is relatively low. Therefore, participating in load response during this period has less impact on the daily life of users. This adjustment helps to optimize energy use and reduce interference with user comfort. In the period from 19:00 to 21:00, although shutting down the load may temporarily affect the comfort of users, due to the high power load and the increase in peak electricity prices at this time, users can obtain larger economic subsidies by participating in load response, so the response power is still relatively large.

[0131] In terms of considering the uncertainty of user response behavior, current research on flexible loads on the user side often only focuses on flexible loads with a single characteristic, and the research on the coordinated scheduling of distributed generation and energy storage in virtual power plants is relatively insufficient. In order to solve these problems, the present invention can minimize the overall operating cost of the virtual power plant by strengthening the coordination between the power supply side and the user side flexible loads.

[0132] The present invention constructs an optimization scheduling model for a virtual power plant that includes traditional power generation, distributed power generation, energy storage, and main-distribution-micro coordination of user loads: considering the uncertainty of user response behavior, the CUCB-Avg algorithm based on the combined multi-armed bandit problem is used to select users, which improves aggregation and accuracy and the range of affected users; a virtual power plant optimization scheduling method is proposed, which takes profit maximization as the goal, significantly improves the scheduling efficiency and economy of the virtual power plant, and optimizes the configuration and use of energy resources. Through precise optimization scheduling, the relationship between traditional power generation and distributed power generation can be better balanced, energy storage equipment can be reasonably used, and adjustments can be made according to the actual situation of user loads, so as to achieve the best resource utilization effect. This can not only improve the overall operating efficiency of the virtual power plant, but also provide strong support for the stable operation of the power system.

[0133] Embodiment 3

[0134] Figure 6 This is a schematic diagram of the structure of a virtual power plant optimization scheduling device provided in the third embodiment of the present invention. Figure 6 As shown, the device comprises:

[0135] The associated data acquisition module 310 is used to acquire associated data corresponding to distributed energy and gas turbine power generation systems in the virtual power plant; wherein the distributed energy includes distributed power sources, energy storage systems and user flexible loads;

[0136] A power plant model building module 320 is used to build a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; wherein the virtual power plant mathematical model includes an operation cost model and an operation income model of the virtual power plant;

[0137] A user load determination module 330 is used to determine a target user load combination participating in the optimal scheduling of the virtual power plant through a multi-armed tiger algorithm for the user flexible load in the virtual power plant;

[0138] The power plant optimization scheduling module 340 is used to determine the optimization scheduling plan of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination.

[0139] The technical solution of the embodiment of the present invention includes obtaining the associated data corresponding to the distributed energy and gas turbine power generation system in the virtual power plant; establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; determining the target user load combination participating in the optimal scheduling of the virtual power plant through a multi-armed slot machine algorithm for the user flexible load in the virtual power plant; and determining the optimal scheduling scheme of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination. In the embodiment of the present invention, in order to solve the technical problem that the uncertainty of the user flexible load response behavior causes the aggregated power of the virtual power plant to deviate from the expected aggregation, affecting the stability of the power system and the economic benefits of the virtual power plant, a multi-armed slot machine algorithm is used to select the optimal target user load combination, thereby improving the aggregation accuracy of the virtual power plant and reducing the scope of affected users. It can not only improve the overall operating efficiency of the virtual power plant, but also provide strong support for the stable operation of the power system.

[0140] Optionally, the power plant model building module 320 includes:

[0141] The operating cost model establishment submodule is used to establish the operating cost model of the virtual power plant based on the operating and depreciation costs of distributed power sources, the cost of purchasing electricity from the power grid, the cost of using fuel for gas turbines, and the compensation costs for optimizing the scheduling of user flexible loads.

[0142] Optionally, the power plant model building module 320 includes:

[0143] The operating revenue model establishment submodule is used to establish the operating revenue model of the virtual power plant based on the operating revenue of the distributed power source corresponding to the virtual power plant, the revenue from selling electricity to the power grid, the power generation revenue of the gas turbine, and the penalty for the user's flexible load not responding or responding incompletely.

[0144] Optionally, the user load determination module 330 includes:

[0145] A first loss value determination submodule, used to determine a reference user load combination participating in the virtual power plant optimization scheduling based on an offline optimal algorithm, and calculate a first loss value corresponding to the reference user load combination;

[0146] A second loss value determination submodule, used to determine a candidate user load combination participating in the virtual power plant optimization scheduling based on the CUCB-Avg algorithm, and calculate a second loss value corresponding to the candidate user load combination;

[0147] The target user load combination determination submodule is used to determine a target regret value based on the first loss value and the second loss value, and when the target regret value meets a preset condition, the candidate user load combination is used as the target user load combination.

[0148] Optionally, the first loss value determination submodule is specifically used for:

[0149] Determining a preset response probability of each user's flexible load in the virtual power plant based on a probability distribution function;

[0150] Determine at least two of the reference user loads from a plurality of user flexible loads based on the preset response probability and the offline optimal algorithm;

[0151] Combining at least two of the reference user loads into the reference user load combination;

[0152] The first loss value is calculated based on the preset response probability and power value corresponding to each of the reference user loads and the target aggregate power value.

[0153] Optionally, the virtual power plant optimization scheduling device further includes: a cyclic update module, which is used to:

[0154] When the target regret value does not meet the preset conditions, the following process is executed cyclically:

[0155] Updating the response probability and the number of selections of each candidate user load in the candidate user load combination;

[0156] Determine a current user load combination based on the updated response probability and the number of selections, and calculate a current loss value corresponding to the current user load combination;

[0157] The second loss value is updated based on the current loss value, and the target regret value is determined based on the first loss value and the updated second loss.

[0158] Optionally, the power plant optimization scheduling module 340 includes:

[0159] An optimization scheduling constraint determination submodule is used to determine the optimization scheduling constraints corresponding to the virtual power plant according to the associated data, wherein the optimization scheduling constraints include distributed generation constraints, power exchange constraints with the power grid, gas turbine generation constraints, load power upper and lower limit constraints and energy storage constraints;

[0160] The submodule for determining the optimal scheduling scheme is used to determine the target cost function corresponding to the virtual power plant based on the mathematical model of the virtual power plant, and to determine the optimal scheduling scheme of the virtual power plant based on the target cost function, the target user load combination and the optimal scheduling constraints.

[0161] Optionally, a scheduling scheme determination submodule is used to:

[0162] According to the target cost function, the target user load combination and the optimized scheduling constraints, the scheduling parameters corresponding to the distributed energy and gas turbine power generation system in the virtual power plant when the value of the target cost function is minimized are determined by a commercial solver.

[0163] The virtual power plant optimization scheduling device provided in the embodiment of the present invention can execute the virtual power plant optimization scheduling method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0164] Embodiment 4

[0165] Figure 7 A schematic diagram of the structure of an electronic device provided for Embodiment 4 of the present invention. 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 may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.), 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 invention described and / or required herein.

[0166] like Figure 7As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0167] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0168] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a virtual power plant optimization scheduling method.

[0169] In some embodiments, the virtual power plant optimization scheduling method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the virtual power plant optimization scheduling method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the virtual power plant optimization scheduling method in any other appropriate manner (e.g., by means of firmware).

[0170] Various implementations of the systems and techniques described above 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 chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including 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.

[0171] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0172] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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), an 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.

[0173] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device 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 a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0174] The systems and techniques described herein may be implemented in a computing system that includes backend 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 frontend 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 backend components, middleware components, or frontend components. The components of the system may 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), a blockchain network, and the Internet.

[0175] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0176] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0177] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A virtual power plant optimization scheduling method, characterized in that: include: Acquire the associated data corresponding to the distributed energy and the gas turbine power generation system in the virtual power plant; wherein the distributed energy includes distributed power sources, energy storage systems and user flexible loads; Establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data; wherein the virtual power plant mathematical model includes an operation cost model and an operation income model of the virtual power plant; For the flexible loads of users in the virtual power plant, a target user load combination participating in the optimal scheduling of the virtual power plant is determined by a multi-armed bandit algorithm; An optimal scheduling scheme for the virtual power plant is determined based on the associated data, the mathematical model of the virtual power plant and the target user load combination.

2. The method according to claim 1, characterized in that The step of establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data includes: An operating cost model of the virtual power plant is established based on the operating and depreciation costs of distributed power sources, the cost of purchasing electricity from the power grid, the cost of fuel used by gas turbines, and the compensation costs for optimizing the scheduling of user flexible loads.

3. The method according to claim 1, characterized in that The step of establishing a virtual power plant mathematical model corresponding to the virtual power plant according to the associated data includes: An operating income model of the virtual power plant is established based on the operating income of the distributed power sources corresponding to the virtual power plant, the income from selling electricity to the power grid, the power generation income of the gas turbine, and the penalty for no response or incomplete response of the user's flexible load.

4. The method according to claim 1, characterized in that: The determining of the target user load combination participating in the optimal scheduling of the virtual power plant by the multi-armed bandit algorithm includes: Determine a benchmark user load combination participating in the optimal scheduling of the virtual power plant based on an offline optimal algorithm, and calculate a first loss value corresponding to the benchmark user load combination; Determine a candidate user load combination participating in the optimal scheduling of the virtual power plant based on the CUCB-Avg algorithm, and calculate a second loss value corresponding to the candidate user load combination; A target regret value is determined based on the first loss value and the second loss value, and when the target regret value meets a preset condition, the candidate user load combination is used as the target user load combination.

5. The method according to claim 4, characterized in that The determining, based on the offline optimal algorithm, a reference user load combination participating in the optimal scheduling of the virtual power plant, and calculating a first loss value corresponding to the reference user load combination includes: Determining a preset response probability of each user's flexible load in the virtual power plant based on a probability distribution function; Determine at least two of the reference user loads from a plurality of user flexible loads based on the preset response probability and the offline optimal algorithm; Combining at least two of the reference user loads into the reference user load combination; The first loss value is calculated based on the preset response probability and power value corresponding to each of the reference user loads and the target aggregate power value.

6. The method according to claim 4, characterized in that The method further comprises: When the target regret value does not meet the preset conditions, the following process is executed cyclically: Updating the response probability and the number of selections of each candidate user load in the candidate user load combination; Determine a current user load combination based on the updated response probability and the number of selections, and calculate a current loss value corresponding to the current user load combination; The second loss value is updated based on the current loss value, and the target regret value is determined based on the first loss value and the updated second loss.

7. The method according to claim 1, characterized in that Determining the optimal dispatching scheme of the virtual power plant according to the associated data, the virtual power plant mathematical model and the target user load combination includes: Determine, according to the associated data, the optimal dispatch constraints corresponding to the virtual power plant, wherein the optimal dispatch constraints include distributed generation constraints, power exchange constraints with the power grid, gas turbine generation constraints, load power upper and lower limit constraints, and energy storage constraints; The target cost function corresponding to the virtual power plant is determined based on the mathematical model of the virtual power plant, and the optimal scheduling scheme of the virtual power plant is determined based on the target cost function, the target user load combination and the optimal scheduling constraint.

8. The method according to claim 7, characterized in that The determining the optimal scheduling scheme of the virtual power plant based on the target cost function, the target user load combination and the optimal scheduling constraint includes: According to the target cost function, the target user load combination and the optimized scheduling constraints, the scheduling parameters corresponding to the distributed energy and gas turbine power generation system in the virtual power plant when the value of the target cost function is minimized are determined by a commercial solver.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the virtual power plant optimization scheduling method described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the virtual power plant optimization scheduling method described in any one of claims 1-8 when executed.

Images