A method for scheduling resources of virtual power plants

Abstract

The present application provides a virtual power plant resource dispatching method, which can be applied to the field of power dispatching technology. The method includes: the virtual power plant calculates the total dispatching demand based on the dispatching demand and the first dispatching margin, wherein the first dispatching margin is calculated by a first model using the dispatching demand as input data; the virtual power plant reports the total dispatching demand to the dispatching platform, and obtains the total dispatching capacity issued by the dispatching platform; the virtual power plant constructs a day-ahead dispatching model based on the dispatching demand, the first dispatching margin and the total dispatching capacity, and uses the day-ahead dispatching model to calculate the first dispatching capacity; the virtual power plant dispatches the first power resource for the user based on the first dispatching capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power dispatching, and in particular to a method for dispatching resources of a virtual power plant. Background Art

[0002] In the context of green development and reforms on the energy supply and consumption sides, virtual power plants that achieve optimal allocation of multiple resources and economic operation based on the aggregation and flexible dispatch of distributed power resources, energy storage, flexible loads, etc. have rapidly generated application demand in China. In the electricity sales market, virtual power plants, as trading parties faced by both the grid side and the user side, buy electricity from the grid side and sell it to the user side. Their main sources of profit are electricity sales revenue and resource optimization and dispatch revenue.

[0003] However, traditional virtual power plant research on resource optimization and scheduling focuses more on the capacity configuration of multiple distributed power resources within the plant. It fails to build a resource optimization and scheduling model for virtual power plants from the user's perspective, making it difficult to mobilize users' enthusiasm in the actual operation of virtual power plants, which is not conducive to the continuous development of demand response. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a virtual power plant resource scheduling method. By constructing a resource optimization scheduling model of the virtual power plant from the user's side, it is possible to mobilize the enthusiasm of users in the actual operation of the virtual power plant, which is conducive to the continuous implementation of demand response.

[0005] In a first aspect, a virtual power plant scheduling resource method is provided, comprising:

[0006] The virtual power plant calculates a total dispatch demand based on the dispatch demand and the first dispatch margin;

[0007] The virtual power plant reports the total dispatch demand to the dispatch platform and obtains the total dispatch capacity issued by the dispatch platform;

[0008] The virtual power plant constructs a day-ahead scheduling model based on the scheduling demand, the first scheduling margin and the total scheduling capacity, and calculates the first scheduling capacity using the day-ahead scheduling model;

[0009] The virtual power plant dispatches the first power resource for the user based on the first dispatch capacity;

[0010] The first scheduling margin is calculated by a first model using the scheduling demand as input data.

[0011] In the above scheme, the total dispatch demand on the user side is calculated using the first dispatch margin that characterizes the user's dispatch potential. Compared with the total dispatch demand directly formed by the sum of the dispatch demands of all users, it has higher flexibility and accuracy, so that the grid-side standby capacity cost can be reduced while meeting the dispatch target. In addition, in the present technical scheme, the first dispatch capacity of the user is calculated by using the day-ahead dispatch model constructed based on the dispatch demand, the first dispatch margin and the total dispatch capacity, so as to realize the construction of a resource dispatch model of the virtual power plant from the user's side, and fully tap the dispatch potential of the user's power generation / power consumption equipment on the basis of comprehensive consideration of the user's power generation capacity or power consumption comfort, so as to mobilize the enthusiasm of the user in the actual operation of the virtual power plant, and form a "win-win-win" situation of the grid side, the virtual power plant and the user side, which is conducive to the continuous development of demand response.

[0012] In some possible implementations, the method further includes:

[0013] The virtual power plant calculates the second dispatch capacity of the user using an intraday dispatch model based on the execution status of the first power resource;

[0014] The virtual power plant dispatches the second power resources for the user based on the second dispatching capacity.

[0015] In some possible implementations, the first model and the intraday scheduling model are respectively at least one of a linear regression model, a logistic regression model, a decision tree, a machine learning model, and a deep learning model.

[0016] In some possible implementations, the virtual power plant constructs a day-ahead scheduling model based on the scheduling demand, the first scheduling margin, and the total scheduling capacity, and calculates the first scheduling capacity using the day-ahead scheduling model, including:

[0017] The virtual power plant divides the users into corresponding user groups based on the first scheduling margin;

[0018] The virtual power plant determines a calculation mode of the first dispatch capacity of the user group;

[0019] The above virtual power plant builds a day-ahead dispatch model based on the above calculation model;

[0020] The virtual power plant obtains the first dispatch capacity of the user group by solving the day-ahead dispatch model;

[0021] The virtual power plant calculates the first dispatching capacity of the user based on the first dispatching capacity of the user group.

[0022] In some possible implementations, the virtual power plant determines the calculation mode of the first dispatching capacity of the user group, including: the virtual power plant determines that the calculation mode of the first dispatching capacity of the user group is the first calculation mode or the second calculation mode based on the integrity of the data and / or the accuracy of the first model, wherein the data includes at least one of the dispatching demand, the parameters of the day-ahead dispatching model, the historical data of the user, and the evaluation results of the user.

[0023] In some possible implementations, the first calculation mode calculates the evaluation result of the user based on a comprehensive evaluation method.

[0024] In some possible implementations, the second computing mode divides users into controllable users and uncontrollable users based on the control rights over the users' power generation / power consumption equipment, wherein the control rights of the controllable users' power generation / power consumption equipment are controlled by the virtual power plant, and the control rights of the uncontrollable users' power generation / power consumption equipment are controlled by the uncontrollable users.

[0025] In some possible implementations, the above-mentioned day-ahead scheduling model includes an objective function and constraints.

[0026] In some possible implementations, the objective function is to minimize the sum of scheduling costs of all user groups, or to maximize the sum of scheduling benefits of all user groups.

[0027] In some possible implementations, the above constraints include virtual power plant output constraints, virtual power plant response capability constraints, first dispatch capacity constraints of user groups, and first dispatch margin constraints of user groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

[0029] Figure 1 It is a flow chart of a virtual power plant dispatching resource provided by this application;

[0030] Figure 2 It is a flow chart of a virtual power plant resource scheduling method provided by the present application;

[0031] Figure 3 It is a flow chart of a step in a virtual power plant scheduling resource method provided in this application. DETAILED DESCRIPTION

[0032] The embodiments of the present invention are described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0033] See also Figure 1, Figure 1 This is a flow chart of a virtual power plant dispatching resource provided in this application.

[0034] The virtual power plant (VPP) aims to aggregate and flexibly regulate distributed power sources, energy storage, flexible loads and other resources. Without changing the existing power grid network topology, it aggregates distributed resources of different types and regions through advanced measurement, communication and control technologies, and achieves coordinated operation among multiple load groups through the coordinated operation of the virtual power plant, thereby reasonably and efficiently dispatching distributed resources in the region and achieving optimal allocation and economical operation of multiple resources.

[0035] like Figure 1 As shown, user 111 generates a dispatching demand 131, user 112 generates a dispatching demand 132, user 113 generates a dispatching demand 133, etc., and uploads them to the virtual power plant 12 respectively. After receiving the dispatching demands 131, 132, 133, etc. uploaded by users 111, 112, 113, etc., the virtual power plant 12 uses the dispatching demands 131, 132, 133, etc. to analyze and obtain the output curve 14 and the total dispatching demand 15, and then reports the output curve 14 and the total dispatching demand 15 to the dispatching platform 16. After receiving the output curve 14 and the total dispatching demand 15 reported by the virtual power plant 12, the dispatching platform 16 uses the output curve 14 and the total dispatching demand 15 to calculate the total dispatching capacity 17 of the virtual power plant 12, and sends the total dispatching capacity 17 to the virtual power plant 12. The virtual power plant 12 receives the total dispatching capacity 17 issued by the dispatching platform 16, and determines the dispatching capacity 181, dispatching capacity 182, dispatching capacity 183... of users 111, user 112, user 113... according to the total dispatching capacity 17, and generates corresponding dispatching instructions 191, dispatching instruction 192, dispatching instruction 193... according to the dispatching capacity 181, dispatching capacity 182, dispatching capacity 183..., and sends them to users 111, user 112, user 113... respectively.

[0036] Based on the aforementioned virtual power plant resource scheduling process, a virtual power plant resource scheduling method provided by the present application is specifically introduced below.

[0037] See also Figure 2 , Figure 2 This is a flow chart of a virtual power plant resource scheduling method provided by this application. Figure 2 As shown, the virtual power plant scheduling resource method provided by the present application includes:

[0038] S201: The virtual power plant obtains scheduling requirements.

[0039] Among them, virtual power plants can be Figure 1 In the virtual power plant 12, users can Figure 1 The scheduling requirements of users 111, 112, 113, etc. can be Figure 1 Scheduling requirements 131, scheduling requirements 132, scheduling requirements 133... in the.

[0040] The dispatching requirements are generated by users and uploaded to the virtual power plant, so that the virtual power plant can obtain the dispatching requirements. Specifically, the virtual power plant invites users to participate in resource scheduling. If the user accepts the invitation to participate in resource scheduling, the dispatching requirements (including price and demand) for each period of power resources will be uploaded to the virtual power plant. Correspondingly, the virtual power plant receives the dispatching requirements uploaded by the user.

[0041] The dispatching demand can be either a power generation dispatching demand or a power consumption dispatching demand. When the dispatching demand is a power generation dispatching demand, the user's equipment is a power generation equipment, such as a power plant boiler, steam turbine, gas turbine, water turbine, generator, transformer, etc. The dispatching capacity is the power generation, and the dispatching instruction is to let the power generation equipment perform power generation behavior. When the dispatching demand is a power consumption dispatching demand, the user's equipment is a power consumption equipment, such as a motor, lighting equipment, electric heating equipment, electrical switches, high and low voltage protection equipment, household appliances (air conditioners, induction cookers), commercial appliances, etc. The dispatching capacity is the power consumption, and the dispatching instruction is to let the power consumption equipment perform power consumption behavior.

[0042] S202: The virtual power plant calculates a total dispatch demand based on the dispatch demand and the first dispatch margin.

[0043] First, the virtual power plant matches the user's dispatch demand with the data in the user database of the virtual power plant, and obtains historical data related to the dispatch demand from the user database. For example, if part of the user's dispatch demand is: electricity consumption location: shopping mall A; electricity consumption time: January 3, 2000; electricity consumption period: 17:00-20:00; planned electricity consumption: 300 kWh, then the data in the user database with the same electricity consumption location, the same electricity consumption time or the same electricity consumption period can be matched as the historical data related to the dispatch demand, such as (1) electricity consumption location: shopping mall A; electricity consumption date: December 25, 1999; electricity consumption period: 14:00-17:00; planned electricity consumption: 150 kWh, actual power consumption: 120 kWh; dispatch margin: 30 kWh; (2) Power consumption location: Shopping Mall B; Power consumption date: January 1, 1999; Power consumption period: 14:00-17:00; Planned power consumption: 130 kWh, actual power consumption: 150 kWh; dispatch margin: -20 kWh; (3) Power consumption location: Shopping Mall C; Power consumption date: December 27, 1999; Power consumption period: 17:00-20:00; Planned power consumption: 230 kWh, actual power consumption: 230 kWh; dispatch margin: 0 kWh.

[0044] Secondly, the virtual power plant takes the user's dispatch demand and related historical data as input data, and uses a linear regression model, a logistic regression model, a decision tree, a machine learning model, a deep learning model or any combination thereof as the first model for prediction, and obtains the first dispatch margin corresponding to the dispatch demand. Among them, the predicted dispatch demand formed by the difference between the dispatch demand and the first dispatch margin can meet the same user's power generation capacity or power comfort compared to the dispatch demand provided by the user. Therefore, the first dispatch margin represents the dispatch potential of the user's power generation / power consumption equipment.

[0045] Next, the virtual power plant takes the dispatching demands of multiple users and the first dispatching margin as input data, and adopts linear regression model, machine learning model, deep learning model or any combination of them as the second model to calculate the total dispatching margin for multiple users during each peak period of the electricity consumption day.

[0046] Finally, based on the market price of electricity resources, the virtual power plant aggregates the dispatching demands of multiple users and the total dispatching margins during each peak period of the electricity consumption day to obtain the output curve and the total dispatching demand (including total price and total demand). The output curve can be Figure 1 In the output curve 14, the total dispatch demand can be Figure 1 The total scheduling demand in is 15.

[0047] S203: The virtual power plant reports the total dispatch demand to the dispatch platform and obtains the total dispatch capacity issued by the dispatch platform.

[0048] The virtual power plant reports the calculated output curve and total dispatch demand to the dispatch platform. Correspondingly, the dispatch platform receives the output curve and total dispatch demand reported by the virtual power plant, calculates the total dispatch capacity of the virtual power plant using the output curve and total dispatch demand, and sends the total dispatch capacity to the virtual power plant. Correspondingly, the virtual power plant receives the total dispatch capacity sent by the dispatch platform, thereby obtaining the total dispatch capacity. The dispatch platform can be Figure 1 In the dispatch platform 16, the total dispatch capacity can be Figure 1 The total dispatch capacity is 17.

[0049] S204: The virtual power plant constructs a day-ahead scheduling model based on the scheduling demand, the first scheduling margin and the total scheduling capacity, and uses the day-ahead scheduling model to calculate the first scheduling capacity.

[0050] In some possible implementations, such as Figure 3 As shown, the calculation method of the first scheduling capacity is as follows:

[0051] S2041: The virtual power plant divides users into corresponding user groups based on the first scheduling margin.

[0052] The virtual power plant places all users into different resource pools according to the first scheduling margin of the users. In a specific implementation, if the first scheduling margin of the user is greater than or equal to the first threshold, the virtual power plant places the user into the first resource pool; if the first scheduling margin of the user is less than the first threshold, the virtual power plant places the user into the second resource pool. In another specific implementation, the virtual power plant places all users into different resource pools with the goal of equal scheduling margin for each resource pool, wherein the scheduling margin of a resource pool is the sum of the first scheduling margins of all users in the resource pool.

[0053] The virtual power plant places all users in the resource pool into different user groups according to the user's response capability. The response capability is the ability to change the power generation / power consumption behavior and adjust the power generation / power consumption equipment according to the first scheduling instruction, which is expressed as the output power difference before and after the power generation / power consumption equipment executes the first scheduling instruction. The output power of the power generation / power consumption equipment before executing the first scheduling instruction can be the first scheduling capacity of the user, and the output power of the power generation / power consumption equipment after executing the first scheduling instruction can be the predicted output power calculated by a linear regression model, a logistic regression model, a decision tree, a machine learning model, a deep learning model or any combination thereof as a third model, and the third model is trained with historical data in the user database as a training set. In a specific implementation, if the user's response capability is greater than or equal to the second threshold, the virtual power plant places the user in the first user group; if the user's first scheduling margin is less than the second threshold, the virtual power plant places the user in the second user group. In another specific implementation, the virtual power plant places all users in different user groups with the goal of equal response capacity of each user group, wherein the response capacity of a user group is the sum of the response capabilities of all users in the user group.

[0054] It should be understood that the above-mentioned method of placing users into resource pools and user groups, as well as the number of resource pools and user groups, is merely an example and is not intended to be a specific limitation.

[0055] S2042: The virtual power plant determines a calculation mode for the first dispatching capacity of the user group.

[0056] In some possible implementations, the virtual power plant determines whether the calculation mode of the first dispatch capacity of the user group is the first calculation mode or the second calculation mode based on the integrity of the data and / or the accuracy of the first model and the accuracy of the second model in the above step S202. The integrity of the data includes the integrity of the parameters of the day-ahead dispatch model, the integrity of the user's historical data, the integrity of the user's dispatch requirements, the integrity of the user's evaluation results, etc. The accuracy of the first model and the accuracy of the second model are calculated based on the historical output data of their respective models.

[0057] In a specific implementation, if the integrity of the data is greater than or equal to a third threshold, that is, the average value of the integrity of data such as parameters of the day-ahead scheduling model owned by the virtual power plant, historical data of users, scheduling requirements of users, and evaluation results of users is greater than or equal to the third threshold, then the first calculation mode is used as the calculation mode for the first scheduling capacity of the user group; if the integrity of the data is less than the third threshold, that is, the average value of the integrity of data such as parameters of the day-ahead scheduling model owned by the virtual power plant, historical data of users, scheduling requirements of users, and evaluation results of users is less than the third threshold, then the second calculation mode is used as the calculation mode for the first scheduling capacity of the user group.

[0058] In a specific implementation, if the average value of the accuracy of the first model and the accuracy of the second model is less than a fourth threshold, the first calculation mode is used as the calculation mode for the first scheduling capacity of the user group; if the average value of the accuracy of the first model and the accuracy of the second model is greater than or equal to the fourth threshold, the second calculation mode is used as the calculation mode for the first scheduling capacity of the user group.

[0059] In a specific implementation, if the integrity of the data is greater than or equal to a third threshold, and the average value of the accuracy of the first model and the accuracy of the second model is less than a fourth threshold, the first calculation mode is used as the calculation mode for the first scheduling capacity of the user group; if the integrity of the data is greater than or equal to the third threshold, and the average value of the accuracy of the first model and the accuracy of the second model is greater than or equal to the fourth threshold, the second calculation mode is used as the calculation mode for the first scheduling capacity of the user group; if the integrity of the data is less than the third threshold, and the average value of the accuracy of the first model and the accuracy of the second model is less than the fourth threshold, the second calculation mode is used as the calculation mode for the first scheduling capacity of the user group; if the integrity of the data is less than the third threshold, and the average value of the accuracy of the first model and the accuracy of the second model is greater than or equal to the fourth threshold, the first calculation mode is used as the calculation mode for the first scheduling capacity of the user group.

[0060] In the first calculation mode, the virtual power plant evaluates the user's electricity use location, dispatching demand, execution effect, response capability and other evaluation indicators based on comprehensive evaluation methods such as hierarchical analysis method, principal component analysis method, data envelopment analysis method, fuzzy evaluation method, etc., obtains the user's evaluation result, and then converts the user's evaluation result into the user's first cost coefficient or first cost. Among them, the better the user's evaluation result, the smaller the user's first cost coefficient or first cost; the worse the user's evaluation result, the larger the user's first cost coefficient or first cost.

[0061] In the second calculation mode, the virtual power plant divides users into controllable users and uncontrollable users based on the control rights over the users' power generation / power consumption equipment. Among them, the control rights of the power generation / power consumption equipment of the controllable users are controlled by the virtual power plant, and the virtual power plant directly controls the operation of the power generation / power consumption equipment according to the first dispatching instruction. The control rights of the power generation / power consumption equipment of the uncontrollable users are controlled by the users, and the virtual power plant issues the first dispatching instruction to the uncontrollable users, and the uncontrollable users control the operation of the power generation / power consumption equipment according to the first dispatching instruction.

[0062] S2043: The virtual power plant builds a day-ahead dispatch model based on the calculation model.

[0063] The day-ahead scheduling model is a linear programming model, which includes an objective function and constraints.

[0064] 1. Objective function

[0065] The objective function is to minimize the sum of the scheduling costs of all user groups or maximize the sum of the scheduling benefits of all user groups.

[0066] (1) When the objective function is to minimize the sum of the scheduling costs of all user groups, the objective function is expressed as:

[0067]

[0068] Where C represents the sum of the scheduling costs of all user groups; represents the scheduling cost of the i-th user group.

[0069] In the first calculation mode, corresponding to the first calculation mode in the above step S2042, when the user's evaluation result is converted into the user's first cost coefficient, the scheduling cost of the i-th user group is expressed as:

[0070]

[0071] c1(t)=·[c3(t)+4(t)+5(t)

[0072] c2(t)=c·[c4(t)+c6(t)]

[0073] in, represents the first dispatch capacity of the i-th user group at time t, c1(t) represents the dispatch price at time t provided by the virtual power plant, represents the first dispatch margin of the i-th user group at time t, c2(t) represents the dispatch subsidy provided by the virtual power plant at time t, c represents the market price of electric power resources, c3(t) represents the benchmark coefficient of electric power resources at time t, c4(t) represents the quality coefficient of electric power resources at time t, c5(t) represents the first cost coefficient of electric power resources at time t, and c6(t) represents the dispatch subsidy coefficient of electric power resources at time t.

[0074] In the first calculation mode, corresponding to the first calculation mode in the above step S2042, when the user's evaluation result is converted into the user's first cost, the scheduling cost of the i-th user group is expressed as:

[0075]

[0076] c1(t)=c·[c3(t)+c4(t)]

[0077] c2(t)=c·[c4(t)+c6(t)]

[0078] in, represents the first dispatch capacity of the i-th user group at time t, c1(t) represents the dispatch price at time t provided by the virtual power plant, represents the first dispatch margin of the i-th user group at time t, c2(t) represents the dispatch subsidy provided by the virtual power plant at time t, represents the first cost of the j-th user in the i-th user group at time t, c represents the market price of electric power resources, c3(t) represents the benchmark coefficient of electric power resources at time t, c4(t) represents the quality coefficient of electric power resources at time t, and c6(t) represents the scheduling subsidy coefficient of electric power resources at time t.

[0079] In the second calculation mode, corresponding to the second calculation mode in the above step S2042, the scheduling cost of the i-th user group is expressed as:

[0080]

[0081] in, represents the first dispatch capacity of the k1th controllable user in the i-th user group at time t, c7(t) represents the dispatch price provided by the virtual power plant for the controllable user at time t, represents the first dispatch margin of the k1th controllable user in the i-th user group at time t, c8(t) represents the dispatch subsidy provided by the virtual power plant to the controllable user at time t, represents the first dispatch capacity of the k2th uncontrollable user in the i-th user group at time t, c9(t) represents the dispatch price provided by the virtual power plant for the uncontrollable user at time t, represents the first scheduling margin of the k2th uncontrollable user in the i-th user group at time t, c 10 (t) represents the dispatch subsidy provided by the virtual power plant to uncontrollable users at time t.

[0082] (2) When the objective function is to maximize the sum of the scheduling benefits of all user groups, the objective function is expressed as:

[0083]

[0084] Among them, Y represents the sum of the scheduling benefits of all user groups; represents the scheduling benefit of the i-th user group.

[0085] In the first calculation mode, corresponding to the first calculation mode in step S2042 above, the scheduling benefit of the i-th user group is expressed as:

[0086]

[0087] c2(t)=c·[c4(t)+c6(t)]

[0088] in, represents the first dispatch margin of the i-th user group at time t, c2(t) represents the dispatch subsidy provided by the virtual power plant at time t, c represents the market price of power resources, c4(t) represents the quality coefficient of power resources at time t, and c6(t) represents the dispatch subsidy coefficient of power resources at time t.

[0089] In the second calculation mode, corresponding to the second calculation mode in the above step S2042, the scheduling benefit of the i-th user group is expressed as:

[0090]

[0091] in, represents the first dispatch margin of the k1th controllable user in the i-th user group at time t, and c8(t) represents the dispatch subsidy provided by the virtual power plant to the controllable user at time t.

[0092] (II) Constraints

[0093] The constraints include: (1) virtual power plant output constraint; (2) virtual power plant response capacity constraint; (3) the first dispatch capacity constraint of the user group; and (4) the first dispatch margin constraint of the user group.

[0094] (1) Virtual power plant output constraints

[0095]

[0096] in, represents the first scheduling capacity of the i-th user group at time t, represents the first scheduling margin of the i-th user group at time t, P VPP (t) represents the first dispatch capacity of the virtual power plant at time t, Represents the total dispatching capacity of the dispatching platform at time t.

[0097] (2) Virtual power plant response capacity constraints

[0098]

[0099]

[0100]

[0101] in, represents the minimum value of the responsiveness of the i-th user group at time t, R set represents the set value of the virtual power plant, R VPP (t) represents the response capability of the virtual power plant at time t, represents the maximum value of the response capability of the i-th user group at time t, represents the first scheduling capacity of the jth user in the i-th user group at time t, represents the maximum value of the predicted output power of the jth user in the i-th user group at time t, It represents the minimum value of the predicted output power of the j-th user in the i-th user group at time t.

[0102] (3) The first scheduling capacity constraint of the user group

[0103]

[0104]

[0105] in, represents the first scheduling capacity of the i-th user group at time t, represents the first scheduling capacity of the jth user in the i-th user group at time t.

[0106] (4) The first scheduling margin constraint of the user group

[0107]

[0108]

[0109] in, represents the minimum value of the first scheduling margin of the jth user in the i-th user group at time t, represents the first scheduling margin of the i-th user group at time t, represents the maximum value of the first scheduling margin of the jth user in the i-th user group at time t, represents the first scheduling capacity of the i-th user group at time t.

[0110] In the second calculation mode, corresponding to the second calculation mode in the above step S2042, the constraints also include: (5) the user's first scheduling capacity constraint; (6) the equipment operation parameter constraint; and (7) the equipment must be turned on or off constraint.

[0111] (5) User's first scheduling capacity constraint

[0112]

[0113] in, represents the minimum value of the first scheduling capacity of the k1-th controllable user in the i-th user group at time t, represents the first scheduling capacity of the k1th controllable user in the i-th user group at time t, It represents the maximum value of the first scheduling capacity of the k1th controllable user in the i-th user group at time t.

[0114] (6) Equipment operating parameter constraints

[0115]

[0116] in, represents the minimum value of the qth operating parameter of the k1th controllable user in the i-th user group at time t, represents the qth operating parameter of the k1th controllable user in the i-th user group at time t, It represents the maximum value of the qth operating parameter of the k1th controllable user in the i-th user group at time t.

[0117] (7) Equipment must be turned on or off

[0118]

[0119]

[0120] in, represents the first scheduling margin of the uncontrollable user k3 in the i-th user group at time t, represents the first scheduling capacity of the uncontrollable user k4 in the i-th user group at time t.

[0121] S2044: The virtual power plant obtains the first dispatch capacity of the user group by solving the day-ahead dispatch model.

[0122] In some possible implementations, the virtual power plant uses a mixed integer linear programming solver to solve the day-ahead dispatch model to obtain the first dispatch capacity for each user group.

[0123] In some possible implementations, the virtual power plant checks the first dispatch capacity of the user group to determine whether the first dispatch capacity of the user group satisfies the constraint conditions in step S2043. If the first dispatch capacity of the user group does not satisfy the constraint conditions in step S2043, the virtual power plant re-solves the day-ahead dispatch model.

[0124] S2045: The virtual power plant calculates the first dispatching capacity of the user based on the first dispatching capacity of the user group.

[0125] In some possible implementations, there may be multiple ways for the virtual power plant to calculate the first dispatch capacity of the user based on the first dispatch capacity of the user group. The first dispatch capacity may be Figure 1 The scheduling capacity 181, scheduling capacity 182, scheduling capacity 183…

[0126] In a specific implementation, the virtual power plant normalizes the dispatching requirements of each user in the user group to obtain a first dispatching coefficient, and then allocates the first dispatching capacity of the user group to each user according to the first dispatching coefficient, thereby obtaining the first dispatching capacity of each user.

[0127] In a specific implementation, the virtual power plant normalizes the first dispatch margin of each user in the user group to obtain a second dispatch coefficient, and then allocates the first dispatch capacity of the user group to each user according to the second dispatch coefficient, thereby obtaining the first dispatch capacity of each user.

[0128] In a specific implementation, the virtual power plant normalizes the response capability of each user in the user group to obtain a third scheduling coefficient, and then allocates the first scheduling capacity of the user group to each user according to the third scheduling coefficient, thereby obtaining the first scheduling capacity of each user.

[0129] It should be understood that the above method of calculating the first scheduling capacity of the user based on the first scheduling capacity of the user group is only an example and is not a specific limitation.

[0130] S205: The virtual power plant dispatches the first power resource for the user based on the first dispatching capacity.

[0131] In some possible implementations, the virtual power plant generates a corresponding first dispatch instruction according to the first dispatch capacity of the user, and dispatches the first power resource for the user according to the first dispatch instruction. The first dispatch instruction may be Figure 1The first power resource is the total dispatch capacity issued by the dispatch platform in step S203.

[0132] In a specific implementation, the virtual power plant selects the first calculation mode in the above step S2042 as the calculation mode of the first dispatching capacity of the user group. After the virtual power plant generates the first dispatching instruction, it sends the first dispatching instruction to all users, and the users obtain the first power resources according to the first dispatching instruction to control the operation of their own power generation / power consumption equipment.

[0133] In another specific implementation, the virtual power plant selects the second calculation mode in the above step S2042 as the calculation mode of the first dispatching capacity of the user group. After the virtual power plant generates the first dispatching instruction, for controllable users, the first power resource is directly obtained according to the first dispatching instruction, and the operation of the power generation / power consumption equipment of the controllable users is controlled; for uncontrollable users, the first dispatching instruction is issued to the uncontrollable users, and the uncontrollable users obtain the first power resource according to the first dispatching instruction and control the operation of their own power generation / power consumption equipment.

[0134] S206: The virtual power plant calculates the second scheduling capacity of the user using the intraday scheduling model based on the execution status of the first power resource.

[0135] When the power generation / power consumption equipment is running on the power consumption day, it sends the real-time operation monitoring data to the virtual power plant. Correspondingly, the virtual power plant receives the real-time operation monitoring data sent by the power generation / power consumption equipment.

[0136] In some possible implementations, the virtual power plant compares the real-time operation monitoring data of the power generation / power consumption equipment with the first dispatch capacity of the corresponding user. If the real-time operation monitoring data of the power generation / power consumption equipment is equal to the first dispatch capacity of the user, it indicates that there is no deviation in the execution of the first power resource; if the real-time operation monitoring data of the power generation / power consumption equipment is not equal to the first dispatch capacity of the user, it indicates that there is a deviation in the execution of the first power resource.

[0137] If there is a deviation in the execution of the first power resource, the virtual power plant calculates the difference between the real-time operation monitoring data and the corresponding first dispatch capacity, and calculates the sum of all differences. Then, the virtual power plant compares the sum of all differences with the fifth threshold. If the sum of all differences is less than or equal to the fifth threshold, the virtual power plant does not generate a second dispatch instruction, and all power generation / power consumption equipment maintains the current operating state; if the sum of all differences is greater than the fifth threshold, the virtual power plant uses the intraday dispatch model to calculate the user's second dispatch capacity.

[0138] The virtual power plant uses the intraday dispatch model to calculate the user's second dispatch capacity, which includes at least the following two specific implementation methods:

[0139] In a specific implementation, the virtual power plant selects the first calculation mode in the above step S2042 as the calculation mode of the first dispatch capacity of the user group, and the virtual power plant uses the sum of the first dispatch margins of all users at the current moment, the second dispatch capacity of all users, and the operation monitoring data of all users as the input data of the intraday dispatch model, thereby calculating the second dispatch capacity of all users at the next moment. Among them, the second dispatch capacity of all users at the current moment is calculated by the intraday dispatch model using the sum of the first dispatch margins of all users at the previous moment, the second dispatch capacity of all users, and the operation monitoring data of all users as input data.

[0140] In another specific implementation, the virtual power plant selects the second calculation mode in the above step S2042 as the calculation mode of the first dispatch capacity of the user group, and the virtual power plant uses the sum of the first dispatch margins of all controllable users at the current moment, the second dispatch capacity of all users (including controllable users and uncontrollable users), and the operation monitoring data of all controllable users as the input data of the intraday dispatch model, thereby calculating the second dispatch capacity of all users (including controllable users and uncontrollable users) at the next moment. Among them, the second dispatch capacity of all users (including controllable users and uncontrollable users) at the current moment is calculated by the prediction model using the sum of the first dispatch margins of all controllable users at the previous moment, the second dispatch capacity of all users (including controllable users and uncontrollable users), and the operation monitoring data of all controllable users as input data.

[0141] In some possible implementations, the intraday scheduling model can be a linear regression model, a logistic regression model, a decision tree, a machine learning model, a deep learning model, or any combination thereof, trained using the historical data of the first scheduling margin, the second scheduling capacity, and the operation monitoring data of the user in the user database as a training set.

[0142] S207: The virtual power plant dispatches a second power resource for the user based on the second dispatch capacity.

[0143] In some possible implementations, the virtual power plant generates a corresponding second dispatch instruction according to the second dispatch capacity of the user, and dispatches a second power resource for the user according to the second dispatch instruction. The second power resource is the sum of the first dispatch margins of all users at the current moment in the above step S206, or the sum of the first dispatch margins of all controllable users at the current moment.

[0144] In a specific implementation, the second power resource is the sum of the first scheduling margins of all users at the current moment in the above step S206. After the virtual power plant generates the second scheduling instruction, it sends the second scheduling instruction to all users, and the users obtain the second power resource according to the second scheduling instruction to control the operation of their own power generation / power consumption equipment.

[0145] In another specific implementation method, the second power resource is the sum of the first dispatching margins of all controllable users at the current moment in the above step S206. After the virtual power plant generates the second dispatching instruction, for controllable users, the second power resource is directly obtained according to the second dispatching instruction, and the operation of the power generation / power consumption equipment of the controllable users is controlled; for uncontrollable users, the second dispatching instruction is issued to the uncontrollable users, and the uncontrollable users obtain the second power resource according to the second dispatching instruction to control the operation of their own power generation / power consumption equipment.

[0146] In summary, by implementing the embodiments of the present invention, compared with the total dispatch demand directly formed by the sum of the dispatch demands of all users, the total dispatch demand on the user side calculated by using the first dispatch margin that characterizes the user's dispatch potential has higher flexibility and accuracy, so that the grid-side standby capacity cost can be reduced while meeting the dispatch target. Moreover, in the present technical solution, the first dispatch capacity of the user is calculated by using the day-ahead dispatch model constructed based on the dispatch demand, the first dispatch margin and the total dispatch capacity, so as to realize the construction of a resource dispatch model of a virtual power plant from the perspective of the user side, and fully tap the dispatch potential of the user's power generation / power consumption equipment on the basis of comprehensively considering the user's power generation capacity or power consumption comfort, so as to mobilize the enthusiasm of the user in the actual operation of the virtual power plant, and form a "win-win-win" situation of the grid side, the virtual power plant and the user side, which is conducive to the continuous development of demand response.

[0147] It should be understood that in the embodiments of the present invention, "when", "when" and "if" all mean that the terminal device or access network device will make corresponding processing under certain objective circumstances, and it is not a limited time, nor does it require the terminal device or access network device to have a judgment action when implementing it, nor does it mean that there are other limitations.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A virtual power plant resource scheduling method, characterized in that: include: The virtual power plant calculates a total dispatch demand based on the dispatch demand and a first dispatch margin, wherein the first dispatch margin is calculated by a first model using the dispatch demand as input data; The virtual power plant reports the total dispatch demand to the dispatch platform, and obtains the total dispatch capacity issued by the dispatch platform; The virtual power plant constructs a day-ahead scheduling model based on the scheduling demand, the first scheduling margin and the total scheduling capacity, and calculates a first scheduling capacity using the day-ahead scheduling model; The virtual power plant dispatches a first power resource for a user based on the first dispatch capacity; The day-ahead scheduling model includes an objective function and constraints; When the objective function is to minimize the sum of the scheduling costs of all user groups, the objective function is expressed as: in, represents the scheduling cost of the i-th user group, expressed as: c1(t)=c·[c3(t)+c4(t)+c5(t)] c2(t)=c·[c4(t)+c6(t)] in, represents the first dispatch capacity of the i-th user group at time t, c1(t) represents the dispatch price at time t provided by the virtual power plant, represents the first dispatch margin of the i-th user group at time t, c2(t) represents the dispatch subsidy provided by the virtual power plant at time t, c represents the market price of power resources, c3(t) represents the benchmark coefficient of power resources at time t, c4(t) represents the quality coefficient of power resources at time t, c5(t) represents the first cost coefficient of power resources at time t, and c6(t) represents the dispatch subsidy coefficient of power resources at time t; Alternatively, the scheduling cost for the i-th user group is expressed as: c1(t)=c·[c3(t)+c4(t)] c2(t)=c·[c4(t)+c6(t)] in, represents the first cost of the jth user in the i-th user group at time t; Alternatively, the scheduling cost for the i-th user group is expressed as: in, represents the first dispatch capacity of the k1th controllable user in the i-th user group at time t, c7(t) represents the dispatch price provided by the virtual power plant for the controllable user at time t, represents the first dispatch margin of the k1th controllable user in the i-th user group at time t, c8(t) represents the dispatch subsidy provided by the virtual power plant to the controllable user at time t, represents the first dispatch capacity of the k2th uncontrollable user in the i-th user group at time t, c9(t) represents the dispatch price provided by the virtual power plant for the uncontrollable user at time t, represents the first scheduling margin of the k2th uncontrollable user in the i-th user group at time t, c 10 (t) represents the dispatch subsidy provided by the virtual power plant to the uncontrollable users at time t; When the objective function is to maximize the sum of the scheduling benefits of all user groups, the objective function is expressed as: in, represents the scheduling benefit of the i-th user group, expressed as: c2(t)=c·[c4(t)+c6(t)] in, represents the first scheduling margin of the i-th user group at time t; Alternatively, the scheduling benefit of the i-th user group is expressed as: in, It represents the first scheduling margin of the k1th controllable user in the i-th user group at time t.

2. The method according to claim 1, characterized in that The method further comprises: The virtual power plant calculates the second dispatch capacity of the user using an intraday dispatch model based on the execution status of the first power resource; The virtual power plant dispatches a second power resource for the user based on the second dispatch capacity.

3. The method according to claim 2, characterized in that The first model and the intraday scheduling model are respectively at least one of a linear regression model, a logistic regression model, a decision tree, a machine learning model, and a deep learning model.

4. The method according to claim 1, characterized in that: The virtual power plant constructs a day-ahead scheduling model based on the scheduling demand, the first scheduling margin, and the total scheduling capacity, and calculates a first scheduling capacity using the day-ahead scheduling model, including: The virtual power plant divides users into corresponding user groups based on the first scheduling margin; The virtual power plant determines a calculation mode of a first dispatch capacity of the user group; The virtual power plant constructs a day-ahead scheduling model based on the calculation mode; The virtual power plant obtains a first dispatch capacity of the user group by solving the day-ahead dispatch model; The virtual power plant calculates the first dispatch capacity of the user based on the first dispatch capacity of the user group.

5. The method according to claim 4, characterized in that The virtual power plant determines a calculation mode of a first dispatch capacity of the user group, including: The virtual power plant determines that the calculation mode of the first scheduling capacity of the user group is the first calculation mode or the second calculation mode based on the integrity of the data and / or the accuracy of the first model, wherein the data includes at least one of the scheduling demand, the parameters of the day-ahead scheduling model, the historical data of the user, and the evaluation results of the user.

6. The method according to claim 5, characterized in that The first calculation mode calculates the evaluation result of the user based on a comprehensive evaluation method.

7. The method according to claim 6, characterized in that The second calculation mode divides users into controllable users and uncontrollable users based on the control rights over the users' power generation / power consumption equipment, wherein the control rights of the controllable users' power generation / power consumption equipment are controlled by the virtual power plant, and the control rights of the uncontrollable users' power generation / power consumption equipment are controlled by the uncontrollable users.

8. The method according to any one of claims 1 to 7, characterized in that: The constraint conditions include a virtual power plant output constraint, a virtual power plant response capability constraint, a first dispatch capacity constraint of a user group, and a first dispatch margin constraint of a user group.

Images