Resource optimization method and control system for virtual power plant

Abstract

The invention relates to the field of smart energy management, and discloses a resource optimization method of a virtual power plant and a control system. The method comprises the steps of selecting a corresponding reference day set according to historical power consumption data of a user and response days, calculating a load average value of each reference day, establishing a user baseline load model, and outputting baseline load data. And based on the baseline load data and the historical power consumption data, in combination with the label information of the user, establishing a multivariate feature-label classification identification model, and generating user portrait information. And based on the user portrait information, comparing user load change data obtained by real-time monitoring with user baseline load data, calculating a load reduction amount or a peak shift amount during a response period, and outputting actual response contribution data of the user. And based on the actual response contribution data of the user and in combination with the virtual power plant income model, calculating a resource compensation limit of the user, and obtaining an income distribution result. Through the method, the problems of insufficient utilization of distributed resources, poor demand response mode flexibility and inaccurate resource heterogeneous characteristic modeling of the virtual power plant can be solved, resource optimization and reasonable income distribution are realized through accurate user baseline load estimation, user characteristic identification and response contribution calculation, and the user experience is improved. And the regulation and control efficiency and economy of the virtual power plant are improved.

Description

Technical Field

[0001] The present application relates to the field of smart energy management, and in particular to a resource optimization method and control system for a virtual power plant. Background Art

[0002] This section is intended to provide a background or context for the embodiments of the present application as stated in the claims. The contents in this section are for reference only and do not constitute an admission or confirmation that they are prior art that has been disclosed.

[0003] In recent years, the power system has faced the dual dilemma of power supply shortages during peak hours and large capacity idleness during off-peak hours. This imbalance between supply and demand has led to a significant increase in power supply costs, a continuous increase in carbon emissions, and increased pressure on grid operations. However, existing power system control technologies are difficult to fully integrate demand-side resources. Traditional methods mainly rely on centralized power generation to balance supply and demand, and insufficiently utilize distributed adjustable loads and energy storage resources, making it impossible to flexibly respond to the regulation needs of peak and off-peak loads.

[0004] In addition, the current demand response model often relies on the regulation capacity of a single resource or region, and fails to achieve efficient aggregation and precise regulation of generalized load resources (such as energy storage, electric vehicles, and distributed power sources). At the same time, the heterogeneous characteristics between different resources (including category differences and parameter differences) lack a unified modeling and optimization method, which makes it difficult for the overall response capacity of virtual power plants to achieve ideal results. This technical defect limits the role of virtual power plants in grid regulation and makes it difficult to meet the requirements of modern power systems for economy, flexibility, and carbon emission reduction.

[0005] To solve the above problems, ideally a virtual power plant with the attributes of a power system is needed, which includes basic resources such as adjustable loads, energy storage and distributed power sources, as well as microgrids and local energy Internets integrated by these basic resources. However, this type of virtual power plant has not yet appeared in my country. Summary of the invention

[0006] The purpose of this application is to provide a resource optimization method and control system for a virtual power plant, which can solve the problems of insufficient utilization of distributed resources in virtual power plants, poor flexibility of demand response modes, and inaccurate modeling of resource heterogeneous characteristics. Through accurate user baseline load estimation, user characteristic identification and response contribution calculation, resource optimization and reasonable distribution of benefits can be achieved, thereby improving the regulation efficiency and economy of virtual power plants.

[0007] The present application discloses a resource optimization method for a virtual power plant, comprising:

[0008] According to the user's historical electricity consumption data, select the corresponding reference day set according to the response day, calculate the load average of each reference day, establish the user's baseline load model, and output the baseline load data;

[0009] Based on the baseline load data and the historical power consumption data, combined with the user's tag information, a multivariate feature-tag classification recognition model is established to generate user portrait information;

[0010] Based on the user portrait information, the user load change data obtained by real-time monitoring is compared with the user baseline load data, the load reduction amount or peak shift amount during the response period is calculated, and the user's actual response contribution data is output;

[0011] Based on the actual response contribution data of the user and combined with the virtual power plant revenue model, the user's resource compensation amount is calculated to obtain the revenue distribution result.

[0012] In a preferred embodiment, the profit distribution model includes:

[0013] According to the user's response contribution C i and the overall revenue S of the virtual power plant, calculate the average marginal contribution of each user in all possible participation orders:

[0014]

[0015] in,

[0016] T is any subset of the virtual power plant that does not contain user i;

[0017] S(T) is the total benefit of subset T;

[0018] |N| is the total number of users of the virtual power plant.

[0019] In a preferred embodiment, the profit distribution model includes:

[0020] The benefits of the virtual power plant are divided equally among the users or according to the preset weights:

[0021]

[0022] in,

[0023] R i is the benefit allocated to user i;

[0024] C i is the load reduction amount of user i.

[0025] In a preferred example, the objects of estimation of the user baseline load include individual users and cluster users;

[0026] The method for estimating the user baseline load includes:

[0027] The response days were divided into weekdays, Saturdays, and Sundays, and baselines were calculated separately;

[0028] If the response day is a working day, the five normal working days before the invitation day are selected to form a baseline calculation reference day set, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the five reference days during the demand response period are calculated;

[0029] If any of the response days is a Saturday, the three Saturdays before the invitation day are selected as the baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated;

[0030] If any of the response days is a Sunday, the three Sundays before the invitation day are selected as baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated.

[0031] In a preferred example, generating user portrait information further includes:

[0032] Acquire the user's load power data, user tag information, and external environment information, remove abnormal data and perform standardization processing on the load power data, and generate standardized load data;

[0033] Based on the standardized load data, feature extraction is performed in the time domain and frequency domain to extract the user's load behavior characteristics, including daily load variation pattern, peak load, valley load, main frequency component and energy distribution, and generate an initial feature set of the user;

[0034] Performing feature selection on the initial user feature set, removing redundant features and noise features, and generating an optimized user feature set;

[0035] Associating the optimized user feature set with user tag information, establishing a feature-tag mapping relationship based on a classification recognition model, and completing a feature-tag classification recognition model through training;

[0036] Using the feature-label classification and recognition model, feature extraction and classification recognition are performed on the load power data of the target user to generate label information of the target user;

[0037] Generate an electricity customer profile based on the tag information of the target user.

[0038] In a preferred embodiment, it also includes:

[0039] Obtaining operating parameters of generalized load resources in the virtual power plant, including category heterogeneous characteristics and parameter heterogeneous characteristics, and generating a resource characteristic data set;

[0040] According to the resource characteristic data set, corresponding mathematical models are established for different types of resources respectively, and a characteristic model of a single resource is output;

[0041] Based on the characteristic model of the single resource, the generalized load resource is subjected to equalization processing to generate an equivalent model of the generalized load resource, and the equivalent model is applied to the grid regulation and operation of the virtual power plant.

[0042] In a preferred example, the baseline load is oriented toward the flexible load in the virtual power plant, and the flexible load is the electricity load that should be consumed assuming that the user does not participate in demand response.

[0043] The present application also discloses a control system for a virtual power plant, comprising:

[0044] A system monitoring module, used to monitor various resource data in the virtual power plant;

[0045] Potential analysis module, used to analyze the potential of the user side, energy storage side and new energy side, and analyze the controllable potential based on the load resource status and load control method;

[0046] The grid interaction module is used to integrate load-side resources to enable network interaction and instant connection between users and between users and the grid;

[0047] A resource optimization module, configured to simulate a basic power simulation service within the virtual power plant through baseline load calculation;

[0048] A dispatching optimization module is used to formulate a day-ahead dispatching optimization method and an intraday dispatching optimization method according to the outputs of the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module, and to feed back the outputs of the day-ahead dispatching optimization method and / or the intraday dispatching optimization method to the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module.

[0049] The present application also discloses a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

[0050] The present application also discloses a computer program product, comprising computer executable instructions, characterized in that when the computer executable instructions are executed by a processor, the steps in the method described in any one of claims 1 to 7 are implemented.

[0051] In the implementation mode of the present application, an accurate assessment of user response contribution is achieved through the establishment of a precise user baseline load model, the generation of user portraits based on feature classification, and the analysis of load changes through real-time monitoring. At the same time, the method combines the virtual power plant revenue model to reasonably allocate resource compensation, thereby solving the problems of inaccurate demand response, insufficient resource regulation flexibility, and lack of basis for revenue distribution in the prior art, and significantly improving the efficiency and economy of virtual power plants in load regulation and resource utilization.

[0052] Furthermore, the revenue distribution model ensures that user revenue is directly linked to actual contribution by introducing a fair distribution method based on marginal contribution, while providing a simple proportional distribution method suitable for rapid distribution scenarios of large-scale users.

[0053] Furthermore, accurate identification of user load behavior and response capability is achieved through accurate estimation of user baseline load and generation of individualized portraits. Flexible selection of reference day sets based on the type of response day, combined with time domain and frequency domain feature extraction methods, not only enhances the accuracy of baseline load estimation, but also provides a reliable data basis for the generation of user portraits, thus providing strong support for the differentiated response strategy and resource scheduling optimization of virtual power plants.

[0054] Furthermore, by modeling and equalizing the heterogeneous characteristics of generalized load resources, complex and diverse resources are integrated into a unified control model, significantly improving the resource integration and grid control capabilities of virtual power plants. At the same time, the clear definition of flexible loads provides a technical basis for the implementation of demand response, effectively supports the application of virtual power plants in load reduction, peak shifting and valley filling, and system optimization, and enhances overall flexibility and practicality.

[0055] Each technical feature disclosed in the above invention content, each technical feature disclosed in each implementation mode and example below, and each technical feature disclosed in the accompanying drawings can be freely combined with each other to form various new technical solutions (these technical solutions should be deemed to have been recorded in this specification), unless such combination of technical features is technically infeasible. For example, in one example, feature A+B+C is disclosed, and in another example, feature A+B+D+E is disclosed, and features C and D are equivalent technical means that play the same role. Technically, only one of them can be used, and it is impossible to use them at the same time. Feature E can be combined with feature C technically. Then, the solution of A+B+C+D should not be deemed to have been recorded because it is technically infeasible, while the solution of A+B+C+E should be deemed to have been recorded. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1It is a flow chart of a resource optimization method for a virtual power plant according to an embodiment of the present application.

[0057] Figure 2 It is a flow chart of a resource optimization method for a virtual power plant according to an embodiment of the present application.

[0058] Figure 3 It is a flow chart of a resource optimization method for a virtual power plant according to an embodiment of the present application.

[0059] Figure 4 It is a flow chart of a resource optimization method for a virtual power plant according to an embodiment of the present application.

[0060] Figure 5 It is a structural diagram of a control system of a virtual power plant according to an embodiment of the present application. DETAILED DESCRIPTION

[0061] In the following description, many technical details are provided to help readers better understand the present application. However, those skilled in the art can understand that the technical solution claimed in the present application can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0062] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below in conjunction with the accompanying drawings.

[0063] The first embodiment of the present application relates to a resource optimization method for a virtual power plant, the process of which is as follows: Figure 1-4 As shown, the method comprises the following steps:

[0064] In S101, according to the historical electricity consumption data of the user, a corresponding reference day set is selected according to the response day, the load average value of each reference day is calculated, a user baseline load model is established, and the baseline load data is output;

[0065] In S102, based on the baseline load data and the historical power consumption data, combined with the user's tag information, a multivariate feature-tag classification recognition model is established to generate user profile information;

[0066] In S103, based on the user profile information, the user load change data obtained by real-time monitoring is compared with the user baseline load data, the load reduction amount or peak shift amount during the response period is calculated, and the user's actual response contribution data is output;

[0067] In S104, based on the user's actual response contribution data and combined with the virtual power plant revenue model, the user's resource compensation amount is calculated to obtain a revenue distribution result.

[0068] In an optional embodiment, the revenue distribution model may include:

[0069] According to the user's response contribution C i and the overall revenue S of the virtual power plant, calculate the average marginal contribution of each user in all possible participation orders:

[0070]

[0071] in,

[0072] T is any subset of the virtual power plant that does not contain user i;

[0073] S(T) is the total benefit of subset T;

[0074] |N| is the total number of users of the virtual power plant.

[0075] In an optional embodiment, the revenue distribution model may include:

[0076] The benefits of the virtual power plant are divided equally among the users or according to the preset weights:

[0077]

[0078] in,

[0079] R i is the benefit allocated to user i;

[0080] C i is the load reduction amount of user i.

[0081] In an optional embodiment, the objects of user baseline load estimation include individual users and cluster users. The individual user baseline load estimation is mainly for a single user, and the estimation result mainly serves the post-settlement of compensation resources between each demand response participant and the virtual power plant; the cluster user baseline load refers to the sum of the user baseline loads of the users aggregated by the virtual power plant, and its estimation result mainly serves the compensation settlement between the virtual power plant and the upper-level system operator.

[0082] In an optional embodiment, the method for estimating the user baseline load includes:

[0083] In S201, the response days are divided into weekdays, Saturdays, and Sundays, and baselines are calculated for each day;

[0084] In S202, if the response day is a working day, the five normal working days before the invitation day are selected to form a baseline calculation reference day set, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the five reference days during the demand response period are calculated;

[0085] In S203, if any of the response days is a Saturday, the three Saturdays before the invitation day are selected as the baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated;

[0086] In S204, if any of the response days is a Sunday, the three Sundays before the invitation day are selected as baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated.

[0087] In an optional embodiment, generating user portrait information may further include:

[0088] In S301, the user's load power data, user tag information and external environment information are obtained, and abnormal data removal and standardization processing are performed on the load power data to generate standardized load data;

[0089] In S302, based on the standardized load data, feature extraction is performed in the time domain and frequency domain to extract the user's load behavior characteristics, including daily load variation pattern, peak load, valley load, main frequency component and energy distribution, and generate the user's initial feature set;

[0090] In S303, feature selection is performed on the initial user feature set to remove redundant features and noise features, and generate an optimized user feature set;

[0091] In S304, the optimized user feature set is associated with the user tag information, a feature-to-tag mapping relationship is established based on the classification recognition model, and the feature-tag classification recognition model is completed through training;

[0092] In S305, a feature-label classification and recognition model is used to extract features and classify the load power data of the target user to generate label information of the target user;

[0093] In S306, a power customer portrait is generated according to the tag information of the target user.

[0094] In general, generalized load resources can include curtailable and transferable loads, electric vehicles, distributed energy storage, etc. Due to the variety of resource types and different operating parameters, generalized load resources present heterogeneous characteristics. At the same time, the same type of load (such as temperature control load) also has loads with different physical properties (such as air conditioners and water heaters), showing parameter heterogeneity.

[0095] In an optional embodiment, the following steps may also be included:

[0096] In S401, the operation parameters of the generalized load resources in the virtual power plant are obtained, including category heterogeneous characteristics and parameter heterogeneous characteristics, and a resource characteristic data set is generated;

[0097] In S402, according to the resource characteristic data set, corresponding mathematical models are established for different types of resources, and a characteristic model of a single resource is output;

[0098] In S403, based on the characteristic model of a single resource, the generalized load resources are processed as equal values ​​to generate an equivalent model of the generalized load resources, and the equivalent model is applied to the grid control and operation of the virtual power plant.

[0099] In an optional embodiment, the baseline load is oriented toward the flexible load in the virtual power plant, and the flexible load is the electricity load that should be consumed assuming that the user does not participate in demand response.

[0100] The second embodiment of the present application relates to a control system of a virtual power plant, whose structure is as follows: Figure 5 As shown in the figure, the control system of the virtual power plant includes:

[0101] The system monitoring module is used to monitor various resource data in the virtual power plant. The system monitoring module includes global monitoring, distributed power monitoring, user-side resource detection and energy storage power station monitoring.

[0102] Potential analysis module, used to analyze the potential of the user side, energy storage side and new energy side, and analyze the controllable potential based on the load resource status and load control method. User side potential analysis: For user side load resources such as residential users, public facilities and electric vehicles, multi-dimensional analysis and sorting of response execution status, response indicators, and response credit index are carried out, the response potential of different load resources is evaluated, and a variety of icons are displayed. The maximum peak-shaving potential of users is displayed, and this data is used as the boundary of the optimization of the user demand response strategy of the virtual power plant to participate in the scheduling optimization calculation. Energy storage side potential analysis: The current status of energy storage, maximum peak-shaving power, maximum power peak-shaving duration, total energy storage configuration capacity, and response speed information of energy storage are displayed, and this data is used as the basic data for scheduling energy storage, participating in the scheduling optimization calculation, and the scheduling results of energy storage are displayed to realize the peak-shaving effect of energy storage. New energy side potential analysis: The power generation status, power generation power, maximum power duration, total configuration capacity, and response speed information of new energy are displayed, and this data is used as the basic data for scheduling the new energy system, participating in the scheduling optimization calculation, and the scheduling results of new energy are displayed.

[0103] The grid interaction module is used to integrate load-side resources to form network interactions and instant connections between users and between users and the grid. Specifically, the grid interaction module can participate in demand response, including invitation or automatic demand response. According to demand response events, the adjustable resource load can be automatically evaluated for each subsystem (energy storage, new energy generation, lighting, HVAC), and regulation strategy recommendations can be formulated. According to manual or automatic methods, internal regulation strategies are implemented to participate in demand response. And the effect of demand response events is evaluated and demand response subsidies are calculated. In addition, the grid interaction module can perform load control, accept superior dispatching and regulation instructions, and decompose the load value of the regulation instruction, generate load control instructions for each subsystem, and realize the real-time automatic response of the virtual power plant to the load regulation instructions of the superior dispatching center.

[0104] The resource optimization module is used to simulate the basic power simulation service within the scope of the virtual power plant through baseline load calculation. The baseline load is calculated for each of the various resources gathered in the virtual power plant, and includes typical day determination, algorithm design, and baseline load correction functions.

[0105] The dispatching optimization module is used to formulate the day-ahead dispatching optimization method and the intraday dispatching optimization method according to the outputs of the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module, and feed back the outputs of the day-ahead dispatching optimization method and / or the intraday dispatching optimization method to the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module.

[0106] In the dispatch optimization, the data of overall power consumption, regional load level, distributed generation, energy storage charging and discharging, response load, etc. are processed and analyzed to solve the predicted operation data of each resource as the output basis of the day-ahead coordinated optimization plan. The adjustable loads of all parties are considered as the optimized dispatch peak-shaving boundary to participate in the dispatch optimization calculation. The day-ahead operation optimization analysis and calculation is to optimize and predict the output of power generation units, power load, energy storage device charging and discharging strategy, controllable load operation adjustment, power grid output and other indicators, and consider the balance of supply and demand to calculate the dispatch optimization results of the virtual power plant for the next day. According to the solved coordination optimization results, the day-ahead coordinated optimization plan of each resource is generated, including the response plan of load-side resources, the adjustable capacity of load-side resources, energy storage charging and discharging plan, power grid purchase plan and other plans. With the coordination optimization plan as the control basis, the day-ahead control dispatch instructions are generated for public resources, controllable user loads, centralized energy storage, etc. Each resource receives the dispatch instruction to generate an operation plan, thereby effectively implementing the coordination optimization plan. Intraday dispatch control is to send and receive control instructions for the actual output of each resource based on the actual output of the distributed power unit, the actual load level and the grid dispatch instruction requirements, and to perform dispatch control. The actual dispatch results are evaluated in real time and fed back to each resource party, including distributed power generation equipment, energy storage equipment and load-side resources, to conduct tracking evaluation and correction of actual operation.

[0107] The first implementation manner is a method implementation manner corresponding to the present implementation manner. The technical details in the first implementation manner can be applied to the present implementation manner, and the technical details in the present implementation manner can also be applied to the first implementation manner.

[0108] Accordingly, the embodiments of the present application also provide a computer-readable storage medium, in which computer executable instructions are stored, and when the computer executable instructions are executed by the processor, the various method embodiments of the present application are implemented. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be a computer-readable instruction, a data structure, a module of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable storage media does not include temporary computer-readable media (transitory media), such as modulated data signals and carriers.

[0109] In addition, an embodiment of the present application further provides a computer program product, which includes computer executable instructions, and when the computer executable instructions are executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0110] It should be noted that, in the present application, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "includes one" does not exclude the presence of other identical elements in the process, method, article or device including the element. In the present application, if it is mentioned that an action is performed according to an element, it means that the action is performed at least according to the element, which includes two situations: performing the action only according to the element, and performing the action according to the element and other elements. Expressions such as multiple, multiple, and multiple include 2, 2 times, 2 kinds, and more than 2, more than 2 times, and more than 2 kinds.

[0111] The serial numbers used in describing the steps of the method do not themselves constitute any limitation on the order of these steps. For example, the step with a larger serial number does not necessarily have to be executed after the step with a smaller serial number. The step with a larger serial number may be executed first and then the step with a smaller serial number. They may also be executed in parallel, as long as this execution order is reasonable for those skilled in the art. For another example, multiple steps with consecutive serial numbers (e.g., step 101, step 102, step 103, etc.) do not limit other steps that can be executed in between. For example, there may be other steps between step 101 and step 102.

[0112] This specification includes combinations of the various embodiments described herein. Individual references to embodiments (e.g., "one embodiment" or "some embodiments" or "preferred embodiments"); however, these embodiments are not mutually exclusive unless indicated as mutually exclusive or it is clear to a person skilled in the art that they are mutually exclusive. It should be noted that the word "or" is used in this specification in a non-exclusive sense unless the context clearly indicates or requires otherwise.

[0113] All documents mentioned in this specification are considered to be included in the disclosure of this application as a whole, so that they can be used as a basis for modification when necessary. In addition, it should be understood that the above is only a preferred embodiment of this specification and is not intended to limit the scope of protection of this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of this specification should be included in the scope of protection of one or more embodiments of this specification.

Claims

1. A resource optimization method for a virtual power plant, characterized in that: include: According to the user's historical electricity consumption data, select the corresponding reference day set according to the response day, calculate the load average of each reference day, establish the user's baseline load model, and output the baseline load data; Based on the baseline load data and the historical power consumption data, combined with the user's tag information, a multivariate feature-tag classification recognition model is established to generate user portrait information; Based on the user portrait information, the user load change data obtained by real-time monitoring is compared with the user baseline load data, the load reduction amount or peak shift amount during the response period is calculated, and the user's actual response contribution data is output; Based on the actual response contribution data of the user and combined with the virtual power plant revenue model, the user's resource compensation amount is calculated to obtain the revenue distribution result.

2. The resource optimization method of a virtual power plant according to claim 1, characterized in that: The revenue distribution model includes: According to the user's response contribution C i and the overall revenue S of the virtual power plant, calculate the average marginal contribution of each user in all possible participation orders: in, T is any subset of the virtual power plant that does not contain user i; S(T) is the total benefit of subset T; |N| is the total number of users of the virtual power plant.

3. The resource optimization method of a virtual power plant according to claim 1, characterized in that: The revenue distribution model includes: The benefits of the virtual power plant are divided equally among the users or according to the preset weights: in, R i is the benefit allocated to user i; C i is the load reduction amount of user i.

4. The resource optimization method of a virtual power plant according to claim 1, characterized in that: The objects of estimation of the user baseline load include individual users and cluster users; The method for estimating the user baseline load includes: The response days were divided into weekdays, Saturdays, and Sundays, and baselines were calculated separately; If the response day is a working day, the five normal working days before the invitation day are selected to form a baseline calculation reference day set, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the five reference days during the demand response period are calculated; If any of the response days is a Saturday, the three Saturdays before the invitation day are selected as the baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated; If any of the response days is a Sunday, the three Sundays before the invitation day are selected as baseline calculation reference days, and the average load Pavi of each reference day during the demand response period and the average load Pavi of the three reference days during the demand response period are calculated.

5. The resource optimization method of a virtual power plant according to claim 1, characterized in that: Generating user portrait information further includes: Acquire the user's load power data, user tag information, and external environment information, remove abnormal data and perform standardization processing on the load power data, and generate standardized load data; Based on the standardized load data, feature extraction is performed in the time domain and frequency domain to extract the user's load behavior characteristics, including daily load variation pattern, peak load, valley load, main frequency component and energy distribution, and generate an initial feature set of the user; Performing feature selection on the initial user feature set, removing redundant features and noise features, and generating an optimized user feature set; Associating the optimized user feature set with user tag information, establishing a feature-tag mapping relationship based on a classification recognition model, and completing a feature-tag classification recognition model through training; Using the feature-label classification and recognition model, feature extraction and classification recognition are performed on the load power data of the target user to generate label information of the target user; Generate an electricity customer profile based on the tag information of the target user.

6. The resource optimization method of a virtual power plant according to claim 1, characterized in that: Also includes: Obtaining operating parameters of generalized load resources in the virtual power plant, including category heterogeneous characteristics and parameter heterogeneous characteristics, and generating a resource characteristic data set; According to the resource characteristic data set, corresponding mathematical models are established for different types of resources respectively, and a characteristic model of a single resource is output; Based on the characteristic model of the single resource, the generalized load resource is subjected to equalization processing to generate an equivalent model of the generalized load resource, and the equivalent model is applied to the grid regulation and operation of the virtual power plant.

7. The resource optimization method of a virtual power plant according to claim 1, characterized in that: The baseline load is oriented toward the flexible load in the virtual power plant, and the flexible load is the electricity load that should be consumed assuming that the user does not participate in demand response.

8. A control system for a virtual power plant, characterized in that: include: A system monitoring module, used to monitor various resource data in the virtual power plant; Potential analysis module, used to analyze the potential of the user side, energy storage side and new energy side, and analyze the controllable potential based on the load resource status and load control method; The grid interaction module is used to integrate load-side resources to enable network interaction and instant connection between users and between users and the grid; A resource optimization module, configured to simulate the basic power simulation service within the virtual power plant through baseline load calculation to obtain the baseline load data as claimed in claim 1; A dispatching optimization module is used to formulate a day-ahead dispatching optimization method and an intraday dispatching optimization method according to the outputs of the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module, and to feed back the outputs of the day-ahead dispatching optimization method and / or the intraday dispatching optimization method to the system monitoring module, the potential analysis module, the power grid interaction module and the resource optimization module.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising computer executable instructions, characterized in that: When the computer executable instructions are executed by a processor, the steps in the method described in any one of claims 1 to 7 are implemented.

Images