Virtual power plant regulation capability analysis method, electronic equipment and storage medium

Abstract

The invention provides a virtual power plant regulation capability analysis method, electronic equipment and a storage medium, which can be applied to the technical field of virtual power plant scheduling and optimization. The method comprises the following steps: determining a response potential factor of a single type of distributed energy resource by using a use characteristic of the single type of distributed energy resource, and determining an adjustable capability limit of the single type of distributed energy resource based on the response potential factor; carrying out aggregation processing on the adjustable capability limits of all types of distributed energy resources to obtain an adjustment capability boundary of the virtual power plant in a multi-scheduling scene; constructing a multi-type resource aggregation interaction capability evaluation optimization model of the virtual power plant, and solving the multi-type resource aggregation interaction capability evaluation optimization model to obtain the multi-type resource aggregation interaction adjustment capability of the virtual power plant; and dynamically analyzing the multi-type resource aggregation interaction adjustment capability of the virtual power plant dynamically changing along with time.

Description

Technical Field

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

[0002] Virtual Power Plant (VPP) is an intelligent management system that integrates and optimizes distributed energy resources (DERs) through information technology and software systems. Its core goal is to participate in the electricity market and grid operation in an aggregated form. The current VPP regulation capacity assessment technology solution mainly relies on independent modeling of DERs, resource response capacity assessment, economic target optimization and market mechanism optimization, predicts the regulation capacity of DERs through historical data and optimization algorithms, and adopts different bidding strategies to improve revenue.

[0003] However, the existing VPP regulation capacity assessment technology solutions have the following main technical problems: (1) There is a lack of a unified quantitative assessment indicator system, which makes it difficult to comprehensively measure the overall regulation capacity of VPP; (2) The timing characteristics and network constraints of DERs are not fully considered, resulting in a large deviation between the assessment results and the actual operation; (3) The optimization method mainly focuses on economic goals and fails to directly reflect the dynamic regulation capacity of VPP; (4) The interactions after DERs aggregation are complex, and existing methods are difficult to effectively model, resulting in insufficient adaptability of scheduling and optimization strategies. Summary of the invention

[0004] In view of the above problems, the present invention provides a virtual power plant regulation capability analysis method, electronic device and storage medium, which are used to solve at least one of the problems in the prior art.

[0005] According to a first aspect of the present invention, a method for analyzing regulation capability of a virtual power plant is provided, comprising:

[0006] Perform multiple unsupervised cluster analyses on the power curves of a single type of distributed energy resources to obtain the usage characteristics of the single type of distributed energy resources, determine the response potential factor of the single type of distributed energy resources using the usage characteristics, and determine the adjustable capacity limit of the single type of distributed energy resources based on the response potential factor;

[0007] Aggregate the adjustable capacity limits of all types of distributed energy resources based on time series to obtain the adjustment capacity boundaries of virtual power plants under different scheduling scenarios;

[0008] Based on the regulation capacity boundary of virtual power plants in different scheduling scenarios, an optimization model for evaluating the multi-type resource aggregation and interaction capacity of virtual power plants is constructed, and the optimization model for evaluating the multi-type resource aggregation and interaction capacity is solved to obtain the multi-type resource aggregation and interaction regulation capacity of virtual power plants.

[0009] Dynamic analysis is conducted on the interactive regulation capabilities of virtual power plants with multiple types of resources that change dynamically over time.

[0010] A second aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0011] The third aspect of the present invention further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the computer program or instructions are executed by a processor.

[0012] The virtual power plant regulation capability analysis method provided by the present invention improves the standardization and accuracy of VPP regulation capability evaluation by constructing a unified quantitative evaluation index system; optimizes resource classification based on a multi-stage clustering method to improve the evaluation accuracy of different types of DERs; adopts a multi-objective optimization algorithm to improve the evaluation reliability of VPP regulation capability, while taking into account economic benefits and operational stability; and combines the entropy weight method to conduct dynamic regulation capability evaluation, so that VPP can more accurately adapt to changes in demand in external scenarios and optimize scheduling strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0014] Figure 1 is an application scenario diagram of a virtual power plant regulation capability analysis method according to an embodiment of the present invention;

[0015] Figure 2 is a flow chart of a method for analyzing the regulation capability of a virtual power plant according to an embodiment of the present invention;

[0016] Figure 3 is an architecture diagram of a method for quantitatively evaluating the regulation capability of a virtual power plant based on multiple stages according to an embodiment of the present invention;

[0017] Figure 4 is a schematic diagram of a theoretical potential assessment model of a single type of distributed energy resources according to an embodiment of the present invention;

[0018] Figure 5is a flow chart of a multi-objective optimization solution method based on AGA-MOPSO and comprehensive membership method according to an embodiment of the present invention;

[0019] Figure 6 It is a block diagram of an electronic device suitable for implementing a virtual power plant regulation capability analysis method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0021] The terms used herein are only for describing specific embodiments, and are not intended to limit the present invention. The terms "including", "comprising", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components. All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used here should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner. In the case of using a statement similar to "at least one of A, B, and C, etc.", it should generally be interpreted according to the meaning of the statement generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to having A alone, having B alone, having C alone, having A and B, having A and C, having B and C, and / or having A, B, C, etc.).

[0022] The present invention belongs to the technical field of virtual power plant (VPP) scheduling and optimization, and involves distributed energy resources (DERs) aggregation scheduling, regulation capability evaluation, demand response optimization and power market interaction. It is specifically applied to smart grids, energy Internet and power market transactions, and is used to enhance the VPP's ability to coordinate and regulate distributed resources, thereby improving the power grid's operating efficiency and market competitiveness.

[0023] The current VPP regulation capacity assessment mainly relies on independent modeling of DERs, resource response capacity assessment, economic target optimization and market mechanism optimization. The regulation capacity of DERs is predicted through historical data and optimization algorithms, and different bidding strategies are used to improve revenue.

[0024] First, existing research has established mathematical models for different types of DERs (such as photovoltaics, wind power, energy storage, electric vehicles, adjustable loads, etc.) to analyze their physical characteristics and regulatory effects on the power grid. For example, photovoltaic and wind power modeling usually uses probabilistic statistical methods or machine learning models to predict the output characteristics of renewable energy, and combines weather, seasons and other factors for evaluation; energy storage system modeling describes the operating characteristics of energy storage through charging and discharging models and optimizes scheduling strategies; demand response (DR) modeling constructs user load behavior models to analyze the flexibility and responsiveness of adjustable loads. However, due to the complex interactions between different DERs, it is difficult for a single model to fully reflect the overall regulatory capabilities of DERs after aggregation, resulting in a lack of integrity and scalability in VPP evaluation.

[0025] Secondly, the existing technologies mainly use statistical analysis methods based on historical data and evaluation methods based on optimization algorithms to evaluate the responsiveness of DERs resources. The statistical analysis method extracts the responsiveness characteristics of DERs through regression models or time series analysis, such as the maximum adjustable power (or adjustable power limit, adjustable power upper limit), response rate and duration, while the optimization algorithm method uses genetic algorithms or dynamic programming to calculate the optimal scheduling scheme and use it to evaluate the overall regulation capability. However, these methods generally rely on historical data, are difficult to adapt to the dynamic changes in the operating status of DERs, and fail to fully consider the synergy between DERs. In addition, the optimization algorithm method usually focuses on economic goals and fails to directly quantify and evaluate the overall regulation capability of VPP.

[0026] Based on the above analysis of existing technical solutions, it can be seen that the existing technologies mainly have the following problems: (1) There is a lack of a unified quantitative evaluation indicator system, which makes it difficult to comprehensively measure the overall regulation capability of VPP; (2) The timing characteristics and network constraints of DERs are not fully considered, resulting in a large deviation between the evaluation results and the actual operation; (3) The optimization method mainly revolves around economic goals and fails to directly reflect the dynamic regulation capability of VPP; (4) The interactions after DERs aggregation are complex, and existing methods are difficult to effectively model, resulting in insufficient adaptability of scheduling and optimization strategies.

[0027] In order to coordinate DERs more effectively, the present invention proposes a quantitative evaluation method for the regulation capacity of virtual power plants based on AI technology. Through the dynamic series connection of machine learning, swarm intelligence and deep learning technology, that is, for the dynamic interactive characteristics and market environment of DERs, a quantitative evaluation method for the regulation capacity of virtual power plants based on AI technology is proposed. With AI as the core engine, the whole process of "feature extraction-aggregation modeling-multi-objective optimization-dynamic evaluation" is opened up, and the refinement level of the evaluation results is significantly improved, providing decision-level support for virtual power plants to participate in multi-type market collaboration, and realizing accurate quantification and dynamic optimization of flexible resource response potential, so as to more effectively coordinate and utilize distributed resources. First, the resource characteristic analysis and evaluation model are combined to more accurately evaluate the flexible resource response potential through qualitative and quantitative analysis; an optimization evaluation model with low operating cost, fast response, large regulation capacity and climbing range as comprehensive goals is established, and considering actual needs, a multi-objective particle swarm optimization algorithm is used to solve the target combination, and the entropy method is used to reflect the dynamic changes of the aggregate response capacity of VPP in the form of scoring. The quantitative evaluation results of the method proposed in the present invention are highly refined, providing an important basis for virtual power plants to participate in multi-type market collaborative optimization.

[0028] In order to better illustrate the advantages of the method provided by the present invention, the method provided by the present invention is described in detail below through specific embodiments and in conjunction with the accompanying drawings.

[0029] Figure 1 It is an application scenario diagram of the virtual power plant regulation capability analysis method according to an embodiment of the present invention.

[0030] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a virtual power plant scheduling and optimization scenario. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0031] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0032] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0033] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0034] It should be noted that the virtual power plant regulation capability analysis method provided in the embodiment of the present invention can generally be executed by the server 105 .

[0035] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0036] The following will be based on Figure 1 The scene described by Figure 2~Figure 5 The virtual power plant regulation capability analysis method of the disclosed embodiment is described in detail.

[0037] Figure 2 4 is a flow chart of a method for analyzing regulation capability of a virtual power plant according to an embodiment of the present invention.

[0038] like Figure 2 As shown, the virtual power plant regulation capability analysis method of this embodiment includes operations S210 to S240.

[0039] In operation S210, multiple unsupervised cluster analyses are performed on the power curve of a single type of distributed energy resource to obtain usage characteristics of the single type of distributed energy resource, and the response potential factor of the single type of distributed energy resource is determined using the usage characteristics, and the adjustable capacity limit of the single type of distributed energy resource is determined based on the response potential factor.

[0040] The above operation S210 is used to perform multiple unsupervised cluster analyses on a single type of resource involved in the virtual power plant. For example, for resource A, first, a cluster analysis is performed on the power curve of a certain user of resource A (technical personnel in the field can use other resource or energy curves according to actual needs), and then, on the premise of obtaining the typical power curves of all users of resource A, the usage characteristics of resource A are determined, and then the response potential factor related to resource A and the adjustable capacity limit of resource A involved in the virtual power plant are obtained; similarly, multi-stage cluster analysis is also performed on other types of resources to obtain the adjustable capacity limit of each type of resource in the virtual power plant.

[0041] In operation S220, the adjustable capacity limits of all types of distributed energy resources are aggregated based on time series to obtain the adjustment capacity boundaries of the virtual power plant under different scheduling scenarios.

[0042] The above operation S210 is used to obtain the adjustable capacity limits of all types of distributed energy resources in the virtual power plant, and then operation S220 is performed to aggregate all types of distributed energy resources in the virtual power plant to obtain the regulation capacity boundaries of the virtual power plant under different scheduling scenarios, that is, the upper and lower limits of the regulation capacity of the virtual power plant under multiple scheduling scenarios.

[0043] In operation S230, a multi-type resource aggregation interaction capability evaluation optimization model of the virtual power plant is constructed based on the regulation capability boundary of the virtual power plant under different scheduling scenarios, and the multi-type resource aggregation interaction capability evaluation optimization model is solved to obtain the multi-type resource aggregation interaction regulation capability of the virtual power plant.

[0044] The above operation S230 is used to obtain the correlation relationship between different types of resources of the virtual power plant, and based on this correlation relationship, more accurately analyze the regulation capability of the virtual power plant.

[0045] In operation S240, a dynamic analysis is performed on the multi-type resource aggregation interactive regulation capability of the virtual power plant that changes dynamically over time.

[0046] The above operation S240 is used to analyze the dynamic changes of the regulation capability of the virtual power plant over time, and based on the analysis results, feedback is given to the multi-resource interactive regulation capability of the virtual power plant and corresponding optimization measures are taken.

[0047] The virtual power plant regulation capability analysis method provided by the present invention improves the standardization and accuracy of VPP regulation capability evaluation by constructing a unified quantitative evaluation index system; optimizes resource classification based on a multi-stage clustering method to improve the evaluation accuracy of different types of DERs; adopts a multi-objective optimization algorithm to improve the evaluation reliability of VPP regulation capability, while taking into account economic benefits and operational stability; and combines the entropy weight method to conduct dynamic regulation capability evaluation, so that VPP can more accurately adapt to changes in demand in external scenarios and optimize scheduling strategies.

[0048] The following is a specific implementation method and combined with the attached Figure 3 The above operations S210 to S240 are further explained in detail.

[0049] Figure 3 It is an architecture diagram of a method for quantitatively evaluating the regulation capability of a virtual power plant based on multiple stages according to an embodiment of the present invention.

[0050] First, if Figure 3 As shown in the figure, AI-enabled feature mining and potential prediction: Based on the K-Means algorithm of unsupervised learning, a theoretical potential assessment of a single type of distributed energy resources is performed. First, cluster analysis is performed on the power curves and energy usage patterns (one of the usage features, the same below) of different types of DERs, their key features are extracted, and their response potential factors are calculated to determine the maximum adjustable capacity of a single type of resource (i.e., the adjustable capacity limit or the upper limit of the adjustable capacity, the same below).

[0051] Secondly, if Figure 3 As shown in the figure, evaluation of the aggregated interactive regulation capability of multiple types of distributed energy resources: according to the characteristics of different types of DERs, an aggregated response capability evaluation model based on time series and optimization methods is constructed to analyze the regulation capability boundary of VPP under different scheduling scenarios.

[0052] Again, if Figure 3 As shown in the figure, AI-enhanced multi-objective global optimization based on the MOPSO algorithm (M‌ulti-‌O‌bjective ‌P‌article ‌S‌warm ‌O‌ptimization) and the Pareto frontier: an optimization evaluation model is established based on the MOPSO algorithm, and a multi-objective evaluation model with low operating cost, fast response and large regulation capacity as the optimization objectives is established. The multi-objective algorithm in artificial intelligence is used, combined with the Pareto optimal solution method, to optimize the overall regulation capability of the VPP.

[0053] Finally, if Figure 3As shown in the figure, dynamic regulation capacity evaluation: the entropy weight method is used to analyze the changing trend of the regulation capacity of DERs, and the deep learning algorithm in artificial intelligence is used to reflect the dynamic changes of the regulation capacity of VPP over time through the scoring mechanism, so as to improve the adaptability and robustness of the evaluation method.

[0054] The following is a specific implementation method and attached Figure 4 Attachment Figure 3 The cluster response potential analysis process involving a single type of distributed energy resources is further described in detail.

[0055] Figure 4 is a schematic diagram of a theoretical potential assessment model for a single type of distributed energy resources according to an embodiment of the present invention.

[0056] like Figure 4 As shown in the figure, a theoretical potential evaluation is conducted for a single type of flexible resource (distributed energy resources, DERs, the same below). First, the K-means clustering method is used to extract the power curve and energy consumption law of distributed energy resources, and five characteristic parameters are extracted. The power of distributed energy resources of the same type is aggregated, and its response potential factor is calculated. The average value of the response potential factor of the power of distributed energy resources of the same type is the response potential factor of the power type of distributed energy resources. It is mainly divided into three links: selection of typical power curves of distributed energy resource power, clustering analysis of typical power curves, and acquisition of demand response potential factors.

[0057] According to an embodiment of the present invention, the above-mentioned multiple unsupervised clustering analyses are performed on the power curve of a single type of distributed energy resource to obtain the usage characteristics of the single type of distributed energy resource, including: performing an initial unsupervised clustering analysis on the power curve of a single user of a single type of distributed energy resource to obtain the typical power curve and energy usage law of the single user of the single type of distributed energy resource; based on the specific characteristics of the selected power curve, performing a secondary unsupervised clustering analysis on the typical power curves of all users of the single type of distributed energy resource to obtain the energy usage pattern of the single type of distributed energy resource, wherein the usage characteristics include energy usage law and energy usage pattern, and the specific characteristics include load rate, daily peak-to-valley difference rate, peak period rate, flat period rate and valley period rate.

[0058] In the above embodiment, cluster analysis is first performed on the daily power curve of a single type of distributed energy resource. According to the response mode of a single type of distributed energy resource, it can be a curve with a year, month, or week as a period. The typical power curve and energy usage law (i.e., one of the usage characteristics, the same below) of a single type of distributed energy resource are obtained, where the energy usage regularity of a single type of distributed energy resource is represented by the clustering upper limit. The higher the clustering system value, the fewer different types of power curves after clustering, the more overlapping the curves are, and the better the clustering effect. The lower the clustering upper limit, the stronger the energy usage law of the resource.

[0059] According to an embodiment of the present invention, the above-mentioned initial unsupervised clustering analysis of the power curve of a single user of a single type of distributed energy resource to obtain the typical power curve and energy consumption law of the single user of a single type of distributed energy resource includes: obtaining the power curve of a single user of a single type of distributed energy resource within a preset period, and based on a preset interval, obtaining the initial clustering points of the power curve of the single user within the preset period; based on the initial clustering points, using the K-means mean clustering method to perform an initial unsupervised clustering analysis on the power curve of the single user within the preset period to obtain an initial clustering result; using the silhouette index method to perform clustering quality evaluation on the initial clustering result, and obtaining the typical power curve and energy consumption law of the single user within the preset period based on the clustering quality evaluation result.

[0060] The following further describes in detail the process of acquiring the above-mentioned typical power curve and energy usage law and the process of clustering analysis of the typical power curve of the equipment through specific implementation methods.

[0061] The typical power curve of a single type of resource can be obtained by calculation. The calculation steps are as follows.

[0062] Step 1: Select the power curve of this type of flexible resource for a certain period , which can usually be a year, a month or a week. The daily power in this period is calculated as one point every 15 minutes, and a total of 96 points are obtained.

[0063] Step 2: Use the k-means algorithm to identify the device The power curves within the period are clustered, and the number of clusters is continuously increased starting from K=2. , until a curve is clustered into one category, the clustering is stopped, and the number of clusters at this time is -1 is the upper limit of resource clustering .

[0064] Step 3: Calculate the number of clusters The following clustering quality evaluation indicators are used. Evaluating the effectiveness of clustering results, i.e. clustering evaluation or validation, is critical to the success of clustering applications. It can ensure that the clustering algorithm identifies meaningful clusters in the data, and can also be used to determine which clustering algorithm is best suited for specific datasets and tasks, and to tune the hyperparameters of these algorithms.

[0065] Since clustering is an unsupervised learning method, there is no ground truth label to compare the clustering results. The minimum amount of constraints satisfied by the Scatter Index (SI) is used as the optimal number of clusters. , the optimal number of clusters The cluster center with the most daily power curves of the equipment is the typical power curve. The calculation method of the silhouette index is shown in formula (1).

[0066] (1),

[0067] in, Indicates the period Vector of daily equipment power curves; Represents the mean value of all daily power curves; Indicates Cluster center vectors; Represents the Euclidean distance between two vectors.

[0068] The process of cluster analysis of typical power curves of equipment is as follows: Cluster analysis is performed on typical power curves of equipment to obtain different operating modes of the equipment (peak-facing type, high load rate type, peak-avoiding type).

[0069] The calculation steps are:

[0070] (1) Select a typical daily power curve of all users. Each daily power curve has one point every 15 minutes, for a total of 96 points.

[0071] (2) Since clustering is to classify resources with the same peak power period every day into one category, if the 96-point daily power curve is selected as the feature vector, there will be a lot of redundant information, which is not conducive to the accuracy of clustering. Therefore, the resource clustering method is used to extract the features. A total of 5 features are selected, namely, the load rate of the curve, the daily peak-to-valley difference rate, the peak period rate, the flat period rate, and the valley period rate. Their definitions and physical meanings are shown in Table 1.

[0072] Table 1: Definition and physical meaning of the five characteristics

[0073]

[0074] (3) Using the five eigenvalues ​​of the resource as the eigenvector, the k-means algorithm is used to cluster the clusters. It is set to 3 categories to obtain the power consumption mode of each resource.

[0075] According to an embodiment of the present invention, the above-mentioned determination of the adjustable capacity limit of a single type of distributed energy resource based on the response potential factor includes: determining the response potential factor of a single type of distributed energy resource based on the energy consumption law and energy consumption mode of a single type of distributed energy resource; obtaining the first response potential of a single type of distributed energy resource based on the response potential factor and active power of a single type of distributed energy resource; determining the second response potential of a single type of distributed energy resource based on the peak-valley power difference of a single type of distributed energy resource; determining the theoretical response potential of a single type of distributed energy resource based on the first response potential and the second response potential of a single type of distributed energy resource; constructing a response potential evaluation model for a single type of resource by introducing a cost incentive target, a peak-valley power difference minimization target and preset constraints, wherein the preset constraints include power reduction constraints, response duration constraints, response time constraints and energy balance constraints; optimizing the theoretical response potential of a single type of distributed energy resource using the response potential evaluation model for a single type of resource to obtain the adjustable capacity limit of a single type of distributed energy resource.

[0076] The following is a further detailed description of the process of obtaining the adjustable capacity limit of a single type of distributed energy resources through a specific implementation method.

[0077] Calculating the adjustable capacity limit of a single type of distributed energy resource (i.e., the maximum adjustable capacity, the same below) involves the theoretical maximum adjustable capacity and the maximum adjustable capacity that takes into account the response cost of the distributed energy resource or the peak-to-valley difference minimization target. First, it is necessary to calculate the response potential factor, and calculate the theoretical maximum adjustable capacity based on the response potential factor and the usage characteristics (parameters of the energy usage law, parameters of the energy usage mode), and then introduce the response cost of the distributed energy resource or the peak-to-valley difference minimization target on the basis of the theoretical maximum adjustable capacity to obtain the actual maximum adjustable capacity.

[0078] Demand response potential factor calculation: Parameters and formulas for demand response potential factor calculation The demand response potential factor is a value between 0 and 1 that indicates the user's demand response potential. 0 means that the resource has no demand response potential at all, and 1 means that the resource can participate in demand response at all times. The calculation of the demand response potential factor mainly involves two parameters: the suitability parameter of the resource and the process / equipment parameter of the resource. After determining the parameters, use formula (2) to calculate the resource's demand response potential factor:

[0079] (2).

[0080] After obtaining the response potential factor, the response resources are determined taking into account the changes in response types. exist The theoretical response potential at the moment is shown in formula (3):

[0081] (3),

[0082] in, Is the response resource exist Active power at the moment, Is the response resource exist Theoretical demand response potential at the time of demand response. Represents the average adjustment when implementing demand response.

[0083] The difference between the peak power and the daily minimum power should not exceed the theoretical potential value of the resource participating in demand response peak load regulation, while the difference between the daily maximum power and the valley power should not exceed the theoretical potential value of valley filling, as shown in formula (4):

[0084] (4),

[0085] in, is the highest or lowest daily power; It has the greatest potential for peak shaving or valley filling; is the duration of the peak or valley; is the number of peak or valley periods.

[0086] Theoretical potential of demand response is the lowest value between the two, as shown in formula (5):

[0087] (5).

[0088] The achievable potential assessment model of a single type of flexible resources includes the construction of an objective function and constraints, as shown in the following steps.

[0089] (1) Objective function

[0090] Compared with previous studies that only considered the single objective of minimizing resource response cost or peak-to-valley difference, this paper considers the objective of minimizing resource response cost or peak-to-valley difference, and constructs a dual objective function that considers both cost and peak-to-valley difference minimization, making the result more accurate. The weight coefficient method is used to simplify the multi-objective optimization problem into a single-objective optimization problem, as shown in formulas (6) to (9):

[0091] (6),

[0092] (7),

[0093] (8),

[0094] (9),

[0095] in, Refers to the maximum power; Refers to the minimum power; the cost of participating in the peak demand response at time t is ; The start and end time of peak demand response are and ; is the objective function with the minimum cost; is the objective function with the minimum peak-to-valley difference; the objective weight coefficients are .

[0096] (10),

[0097] In formula (10), Indicates user exist The amount of power increase or decrease at any moment; Represents the unit power cost.

[0098] (2) Constraints

[0099] 1. Power reduction constraints

[0100] The maximum power reduction amount of each resource type for peak shaving demand response is not higher than the theoretical response potential of the resource, as shown in formula (11):

[0101] (11),

[0102] in, represents the theoretical maximum response potential; Indicates whether there is a peak shaving demand response at time t. , when no response .

[0103] 2. Response duration constraints

[0104] The response duration constraint must be introduced because some devices cannot respond to changes in the device operating environment, and the recovery time will be extended. Since some devices may have a longer response time when facing changes in the device operating environment, their response performance may be affected to a certain extent, so the response duration constraint needs to be considered, as shown in formula (12):

[0105] (12),

[0106] in, Indicates the maximum daily power reduction time.

[0107] 3. Response time constraints

[0108] Since the power consumption of resources varies, the response requirements may not be fully met. This is due to the energy consumption pattern of resources, the fluctuation of demand, and the supply capacity of the system. It cannot be guaranteed that all response requirements can be met in time at all times. Therefore, the response time is subject to certain constraints, as shown in formula (13):

[0109] (13),

[0110] The time when peak shaving demand response cannot be performed is express.

[0111] 4. Energy balance constraints

[0112] After the adjustable load resources participate in the demand response, the curve changes to a certain extent. However, the sum of the load adjustment amount cannot be greater than 0, that is, the total load value cannot increase after participating in the demand response, otherwise it cannot meet the energy balance demand, as shown in formula (14):

[0113] (14).

[0114] if When the load decreases, ;if When the load increases, .here Indicates the direction of load transfer, Indicates that the load is The amount of load transferred at any time.

[0115] Before building a multi-type resource aggregation interaction capability evaluation optimization model and calculating the multi-type resource aggregation interaction regulation capability of a virtual power plant, it is necessary to evaluate and analyze the multi-type resource aggregation interaction regulation capability. Figure 3 The multi-type resource aggregation interactive regulation capability involved in the above operation S220 (i.e., the regulation capability boundary of the virtual power plant in different scheduling scenarios, the same below) is further explained in detail.

[0116] like Figure 3 As shown, the second stage involved in the virtual power plant regulation capacity analysis method is the evaluation and analysis of the interactive regulation capacity of multi-type resource aggregation.

[0117] In order to obtain the multi-index characteristics of the virtual power plant, it is necessary to combine the "aggregation characteristics" of the virtual power plant formed by aggregating different types of flexible resources with the aggregation regulation capability of the VPP resources. By determining key indicators such as baseline power, upper and lower boundaries of regulation capacity, ramp rate limit, and response time range, it is ensured that the capacity of each resource is fully utilized and an accurate output plan is generated while meeting grid stability and market demand.

[0118] After the flexible resource energy consumption characteristics are extracted, they are aggregated according to the characteristic parameters to characterize the aggregate characteristics of multi-type resource aggregates, as shown in formula (15):

[0119] (15),

[0120] in, , , , Respectively represent resource types Always The upper and lower bounds of instantaneous power and the upper and lower bounds of cumulative power consumption, An abbreviation for the access resource type, including conventional loads (power consumption is generally unchangeable, such as lighting loads), electric vehicle charging loads, distributed new energy resources, conventional units in the system, etc.

[0121] For the evaluation of the interactive regulation capacity of multi-type resource aggregation, the time scale is first refined, and then the demand response potential of each industry is superimposed at a single time point to improve the accuracy of the multi-resource demand response potential results. The calculation formula is shown in formula (16):

[0122] (16),

[0123] Right now represents the actual demand response potential of multiple resources at time t, and Representative industries actual demand response potential.

[0124] According to an embodiment of the present invention, the above-mentioned multi-type resource aggregation interaction capability evaluation optimization model for a virtual power plant based on the regulation capability boundary of the virtual power plant in different scheduling scenarios includes: performing a baseline prediction of the active power of the virtual power plant based on the objective function of the lowest total operating cost, the electric power balance constraint and the equipment constraint to obtain the power curve of the virtual power plant when no demand response occurs; using the power curve of the virtual power plant when no demand response occurs and the regulation capability boundary in different scheduling scenarios to obtain the regulation capacity objective function, the ramp rate objective function and the response time objective function of the multi-type resource aggregation interaction capability evaluation optimization model; using the operation constraints and state constraints of all types of distributed energy resources as the constraints of the multi-type resource aggregation interaction capability evaluation optimization model, wherein the constraints of the multi-type resource aggregation interaction capability evaluation optimization model include the ramp rate constraint, the power output constraint, the capacity state constraint, the capacity limit constraint, the response power constraint and the response duration constraint.

[0125] The above embodiments involve Figure 3 The evaluation of the interactive capability of multiple types of resources in VPP mainly includes VPP distributed energy resources (i.e. Figure 3 The flexible resources shown in the figure are the same below) baseline prediction, VPP multi-type resource aggregation and interaction capability evaluation optimization model (i.e. Figure 3 Multi-type resource aggregation potential assessment model shown).

[0126] Baseline prediction of flexible resource operation of virtual power plant: Baseline power refers to the overall power flow curve of VPP without adjusting response demand. It is calculated under the goal of minimizing total operating cost. The corresponding objective function is expressed as shown in formulas (17) and (18):

[0127] (17),

[0128] (18),

[0129] in, A series of operating moments, the set of which can be determined according to actual operating requirements; for The operating cost function of VPP at the moment; Covers the total set of all equipment in the VPP except fixed loads; each equipment has a unique equipment number to identify; For the Resources in Output power at all times (the load output power can be adjusted to take negative values); is the total power of fixed non-adjustable loads in the VPP; for The power purchase and sale of VPP at the moment, with the power purchase as positive; define the Devices The cost function at time ; At the same time, the VPP's electricity purchase and sales price at that time is also included in the calculation.

[0130] The current output power of VPP is shown in formula (19):

[0131] (19).

[0132] When optimizing the scheduling of VPP, in addition to considering the equipment constraints of DERs, it is also necessary to ensure the balance of electric power, as shown in formula (20):

[0133] (20).

[0134] The optimization model for evaluating the interactive capability of multi-type resource aggregation in virtual power plants involves objective functions and constraints, where the objective function includes upper and lower limits of regulation capacity, upper and lower limits of ramp rate, response time range, etc.

[0135] (1) Adjust the upper and lower limits of capacity

[0136] In the optimization model, the upper and lower limits of the regulation capacity can be indirectly determined by calculating the maximum and minimum power of electricity purchase and sale. , , that is, the upper and lower limits of the power purchase and sale reflect the extreme values ​​of the output power range that the VPP can achieve at a specific time. The corresponding objective function is expressed as shown in formulas (21) and (22):

[0137] (twenty one),

[0138] (twenty two).

[0139] In addition, due to network restrictions, the maximum and minimum power constraints of electricity purchase and sale must also be considered, as shown in formula (23):

[0140] (twenty three).

[0141] (2) Upper and lower limits of climbing rate

[0142] Climbing rate limit , It refers to the maximum increase or decrease in output power allowed from one moment to the next when the VPP is operating at its power baseline. The corresponding objective function can be described as shown in formulas (24) and (25):

[0143] (twenty four),

[0144] (25).

[0145] (3) Response time range

[0146] VPP response time Defined as the time from the moment an adjustment command is received to the moment when all regulation resources jointly adjust the total output power to the specified change value while following the power baseline operation The shortest time required. The corresponding objective function is shown in formula (26):

[0147] (26),

[0148] Among them, the collection By equally spaced composition; are the various time points considered in the summation operation; is a binary variable indicating whether the Adjustment instructions Furthermore, when the instruction requirements are not met, the following two constraints must be followed, as shown in formulas (27) and (28):

[0149] (27),

[0150] (28).

[0151] Formulas (27) and (28) indicate that once the requirements of the regulation instruction are met at a certain moment, the subsequent moments will no longer be counted as response time.

[0152] (4) Other constraints

[0153] In order to further obtain the optimization model of aggregated responsiveness, according to the independent response capability models of various DERs, the operation constraints and state constraints of DERs are considered as part of the constraints of the VPP aggregated regulation responsiveness optimization model.

[0154] Distributed generation, load resources and energy storage resources must comply with the ramp rate constraint, as shown in formula (29):

[0155] (29).

[0156] Distributed power sources and energy storage resources must comply with power output constraints, as shown in formula (30):

[0157] (30).

[0158] Energy storage resources must comply with capacity state and capacity limit constraints, as shown in equations (31) and (32):

[0159] (31),

[0160] (32).

[0161] The load resources must also meet the response power constraint and response duration constraint, as shown in formula (33):

[0162] (33).

[0163] According to an embodiment of the present invention, the above-mentioned solution of the multi-type resource aggregation interaction capability evaluation optimization model to obtain the multi-type resource aggregation interaction regulation capability of the virtual power plant includes: performing weighted operation on the objective function of the multi-type resource aggregation interaction capability evaluation optimization model to obtain the joint objective function of the multi-type resource aggregation interaction capability evaluation optimization model; initializing the weight parameters used by the adaptive grid-based multi-objective ion swarm optimization algorithm and the required number of iterations, and generating the initial population of the adaptive grid-based multi-objective ion swarm optimization algorithm based on the joint objective function; obtaining the fitness of particles in the initial population or the current iterative population based on the joint objective function. value, and determine the dominance relationship between particles in the initial population to obtain the Pareto solution set in the current iteration round; use the multi-objective ion swarm optimization algorithm based on adaptive grid to dynamically adjust the distance between particles in the initial population or the current iteration population to obtain an updated Pareto solution set; iteratively perform fitness value acquisition operations, Pareto solution set acquisition operations, and population ion dynamic adjustment operations until the number of iterations is met, and use the linear membership function to obtain the comprehensive membership of all solutions in the updated Pareto solution set; the solution with the largest comprehensive membership in the updated Pareto solution set is used as the model target solution that characterizes the multi-type resource aggregation and interactive regulation capabilities of the virtual power plant.

[0164] The above embodiments relate to the Figure 3 The implementation process of the virtual power plant interactive capability evaluation considering multiple time scales is shown in the following. Figure 5 The above process is further explained in detail.

[0165] Figure 5 It is a flow chart of a multi-objective optimization solution method based on AGA-MOPSO and comprehensive membership method according to an embodiment of the present invention.

[0166] The VPP interactive regulation capability evaluation optimization model established in the present invention involves multiple constraints, including power balance relationship, power purchase and sale constraints, resource call constraints, resource physical operation characteristic constraints and other factors, aiming to optimize multiple objective functions such as minimum total cost, shortest response time, maximum regulation capacity and ramp rate. Therefore, the evaluation of aggregate regulation capability is a dynamic rolling, multi-variable and multi-constrained optimization problem. Considering the relevant series of constraints, solving the objective function formula (i.e., formula (17)) in the optimization model and extracting the optimal decision variables at each moment can obtain the operation baseline. Based on the baseline power, formulas (20)-(22), (24), (25) and (26) are respectively obtained to obtain the corresponding other aggregate regulation response capabilities, which can be reported to the operation control center after combination.

[0167] The method of solving each objective function separately does not accurately describe the feasible adjustment range of VPP. For example, the solution of the response time objective function in some cases may be greater than In this case, it is considered that the current regulation demand VPP cannot participate in the response. To avoid such a situation, the upper and lower limits of regulation capacity, upper and lower limits of ramping, and response time range can be weighted and jointly solved to meet the needs of actual demand, as shown in formula (34):

[0168] (34),

[0169] in, , , , , and is the weight of each objective function. From formula (34), it can be seen that when the VPP operation control center adjusts the demand preference of VPP performance to adapt to various application scenarios, the weight coefficient of each objective can be changed to obtain different solution results of the adjustment capacity range of VPP. There are many strategies to find the best solution for this kind of multi-objective optimization problem.

[0170] The present invention adopts the Pareto frontier and comprehensive evaluation strategy, that is, firstly, the Pareto solution set is generated by using the Multi-objective Particle Swarm Optimization based on AdaptiveGrid Algorithms (AGA-MOPSO) based on adaptive grid. By adaptively adjusting the search strategy of particles, the solution space is effectively explored to generate a set of non-inferior solutions with wide coverage and uniform distribution. Subsequently, the comprehensive membership method is used to screen out the comprehensive optimal solution from the Pareto solution set. This method evaluates the relative advantages and disadvantages of each solution by calculating its membership score, and then selects the final optimal solution in combination with the specific target preferences of the decision maker.

[0171] The present invention selects a linear membership function to calculate the comprehensive membership of each solution in the Pareto solution set. The solution with the largest comprehensive membership is the comprehensive optimal solution, and the corresponding VPP benchmark and boundary capacity at each time form the optimal evaluation solution. The objective function, The membership degree of a solution is shown in formula (35):

[0172] (35),

[0173] in, For the The solution for The membership degree of the objective function is larger, and the larger the value is, the better the solution is. The better the objective function is; , They are the first The set of objective function values The maximum and minimum values ​​in ; For the The solution The objective function value.

[0174] No. The comprehensive membership of the solution As shown in formula (36):

[0175] (36),

[0176] in, is the total number of objective functions; For the The weight of an objective function in all objective functions; is the total number of solutions in the Pareto solution set.

[0177] The process of multi-objective optimization method is as follows Figure 5 As shown in the figure is the number of iterations, is the fitness function of the particle, , , For the The particle swarm, Pareto solution set and objective function solution set at the iteration. Based on this flowchart, the model solution steps can be obtained.

[0178] Step 1: Basic parameter input: Input the operating status information of the VPP resources, and calculate and upload their regulation capacity and expected cost based on the regulation characteristic model.

[0179] Step 2: Population initialization. Set the population size, number of iterations and inertia weight parameters of the optimization algorithm. Generate the initial population according to the number of objective functions. , and ensure that the particles satisfy the feasibility of the system constraints.

[0180] Step 3: Fitness evaluation. Calculate the fitness value of the particle according to the model target , determine its dominance relationship to obtain the Pareto frontier .

[0181] Step 4: Adaptive grid loop iteration. Introduce an adaptive grid mechanism to optimize the particle swarm algorithm, estimate particle density, dynamically adjust crowding distance, and update particle speed and position; then determine whether the maximum number of iterations has been reached, and if so, output the Pareto solution set. .

[0182] Step 5: Optimal solution selection. Combined with the objective function weight input , the comprehensive membership function method is used to determine the comprehensive membership of each objective function , from the solution set Select the optimal solution to obtain the objective function value , and obtain the optimal evaluation result considering the trade-offs among multiple objectives.

[0183] According to an embodiment of the present invention, the above-mentioned dynamic analysis of the multi-type resource aggregation interactive regulation capability of the virtual power plant that changes dynamically over time includes: data standardization of the regulation capacity, ramp rate and response time of the virtual power plant that changes dynamically over time, and obtaining the information entropy and redundancy of the regulation capacity, ramp rate and response time after data standardization; using the information entropy and redundancy of the information entropy to obtain the weight coefficients of the regulation capacity, ramp rate and response time after data standardization; averaging the regulation capacity, ramp rate and response time after data standardization, and respectively calculating the averaging-processed regulation capacity, ramp rate and response time with the corresponding weight coefficients to obtain dynamic analysis results of the regulation capacity, ramp rate and response time.

[0184] The above embodiments relate to the Figure 3 Evaluation of the aggregated interactive regulation capability of virtual power plants based on the entropy weight method: The weights of the above multi-objective solution method are formulated by decision makers based on their experience or preferences, which to a certain extent adapts to different actual business needs. However, in order to more objectively and directly reflect the regulation capability of virtual power plants, weights can be assigned by the entropy weight method to quantify each capability indicator. This weight allocation takes into account the stability of the regulation capability indicators of virtual power plants in various time periods, that is, the greater the fluctuation of the regulation capability indicators, the lower the corresponding indicator weight should be. In order to objectively allocate weights, a method based on information entropy is adopted to determine the weights based on the volatility of the indicators. A lower information entropy means that the indicator changes greatly, so the amount of information is large, and its weight in the comprehensive evaluation should be higher. Conversely, a higher information entropy means that the indicator changes little, the amount of information is small, and the weight should be relatively low. The weighting steps of the entropy weight method are as follows:

[0185] (1) Data standardization, calculate the weight of each indicator in all evaluation objects. Assume that Indicators ,in . The standardized value of the index data is , the standardized calculation is shown in formula (37):

[0186] (37).

[0187] (2) Calculate the information entropy of each indicator, and use the standardized indicator value to calculate the information entropy of each indicator, and then obtain the redundancy of the information entropy. According to the definition of information entropy in information theory, the information entropy of a set of data is shown in formulas (38) and (39):

[0188] (38),

[0189] (39),

[0190] like , then define .

[0191] (3) Determine the weight of each indicator. Determine the weight of each indicator according to its redundancy. Indicators with low redundancy (low information entropy) are given higher weights. According to the information entropy calculation formula, calculate the information entropy of each indicator. . Calculate the weight of each indicator through information entropy , as shown in formula (40):

[0192] (40).

[0193] (4) The evaluation values ​​of each indicator are shown in formula (41):

[0194] (41).

[0195] (5) The value score of the virtual power plant aggregation regulation capability is shown in formula (42):

[0196] (42).

[0197] Compared with the prior art, the technical solution of the present invention has the following advantages:

[0198] (1) The present invention establishes a universal quantitative evaluation index system to quantitatively analyze the typical resource adjustable capacity under different power generation mechanisms and user behaviors. The system comprehensively considers the regulation capacity, timing characteristics and market environment of distributed energy resources (DERs), and uses the entropy method to weight each indicator to ensure the objectivity and comparability of the evaluation results, thereby overcoming the problem of single evaluation indicators in the prior art and difficulty in comprehensively measuring the overall regulation capacity of virtual power plants (VPPs), and improving the accuracy and applicability of the evaluation.

[0199] (2) This invention proposes a source-load resource response capability assessment method based on two-stage flexible resource clustering, which combines DERs resource characteristic analysis with the assessment model to achieve fine classification of resources, and evaluates the response potential of resources based on the classification results. The first stage performs cluster response potential analysis on a single type of flexible resource, and the second stage evaluates the interactive regulation capability of multi-type resource aggregation. When evaluating the demand response potential of flexible resources, current research focuses on modeling and analysis based on load characteristics, while ignoring other characteristics such as response cycle and response duration. Qualitative and quantitative research are two types of existing research. The conclusions of qualitative research often compare the demand response potential of various flexible resources, and it is impossible to obtain an exact value; quantitative research makes up for the above shortcomings, but the result is always a number, that is, the demand response potential for the whole day adopts a fixed value, which is unreasonable, and the single objective function will cause the result to differ greatly from reality. In addition, when examining the demand response potential of regional multi-load aggregation, the result will have a higher error due to this fault. Therefore, the present invention comprehensively considers the characteristics of resources and constructs a two-stage demand response potential evaluation model based on "single achievable potential-multi-type resource potential" to make resource classification more accurate. This method overcomes the problem that the existing technology does not fully consider the dynamic characteristics of DERs and network constraints, enhances the adaptability of the evaluation model, and improves the utilization efficiency of flexible resources.

[0200] (3) The present invention establishes an optimization evaluation model with the goals of low operating cost, fast response and large regulation capacity. The model comprehensively considers key indicators such as the operating cost, response speed, regulation capacity and climbing range of the VPP, and adopts a multi-objective particle swarm optimization algorithm to solve it. By introducing the concept of Pareto optimal solution, different objectives are weighed and optimized to maximize the regulation capacity of the VPP while ensuring the optimal economic benefits, thus overcoming the problem that the existing optimization methods only focus on economic goals and fail to effectively evaluate the overall regulation capacity of the VPP, and achieving a balanced optimization of economic benefits and regulation capacity.

[0201] (4) The present invention adopts the entropy method to dynamically reflect the changes in the aggregate response capability of VPPs. By calculating the score of the aggregate response capability of VPPs, the change trend of the VPP regulation capability under different market environments and load conditions is dynamically reflected. The entropy method can adaptively adjust the weight of each indicator to make the score more representative. At the same time, combined with the time series analysis method, the VPP regulation capability is tracked and optimized over a long period of time, thereby overcoming the problem that the existing methods are difficult to directly quantify the overall regulation capability of VPPs, and improving the flexibility and competitiveness of VPPs under different market environments.

[0202] Figure 6 It is a block diagram of an electronic device suitable for implementing a virtual power plant regulation capability analysis method according to an embodiment of the present invention.

[0203] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0204] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present invention by executing the program stored in the one or more memories.

[0205] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0206] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0207] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0208] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0209] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention may be combined and / or combined in various ways. All of these combinations and / or combinations fall within the scope of the present invention.

[0210] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination advantageously. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A virtual power plant regulation capacity analysis method, characterized in that: The method comprises: Performing multiple unsupervised cluster analyses on power curves of a single type of distributed energy resources to obtain usage characteristics of the single type of distributed energy resources, determining a response potential factor of the single type of distributed energy resources using the usage characteristics, and determining an adjustable capacity limit of the single type of distributed energy resources based on the response potential factor; Aggregate the adjustable capacity limits of all types of distributed energy resources based on time series to obtain the adjustment capacity boundaries of virtual power plants under different scheduling scenarios; Based on the regulation capability boundary of the virtual power plant in different scheduling scenarios, a multi-type resource aggregation interaction capability evaluation optimization model of the virtual power plant is constructed, and the multi-type resource aggregation interaction capability evaluation optimization model is solved to obtain the multi-type resource aggregation interaction regulation capability of the virtual power plant; A dynamic analysis is performed on the virtual power plant's ability to aggregate and interactively regulate multiple types of resources that change dynamically over time.

2. The method according to claim 1, characterized in that Multiple unsupervised clustering analyses are performed on the power curves of a single type of distributed energy resources to obtain the following usage characteristics of the single type of distributed energy resources: Performing a primary unsupervised cluster analysis on the power curve of a single user of the single type of distributed energy resource to obtain a typical power curve and energy usage pattern of the single user of the single type of distributed energy resource; Based on the specific characteristics of the selected power curve, a secondary unsupervised clustering analysis is performed on the typical power curves of all users of the single type of distributed energy resources, and the energy usage pattern of the single type of distributed energy resources, wherein the usage characteristics include energy usage rules and energy usage patterns, and the specific characteristics include load rate, daily peak-to-valley difference rate, peak period rate, flat period rate and valley period rate.

3. The method according to claim 2, characterized in that The power curve of a single user of the single type of distributed energy resource is subjected to an initial unsupervised cluster analysis to obtain a typical power curve and energy usage law of a single user of the single type of distributed energy resource, including: Obtaining a power curve of a single user of the single type of distributed energy resources within a preset period, and obtaining an initial clustering point of the power curve of the single user within the preset period based on a preset interval; Based on the initial clustering points, an initial unsupervised clustering analysis is performed on the power curve of a single user within the preset period using a K-means mean clustering method to obtain an initial clustering result; The initial clustering result is evaluated for clustering quality using a silhouette index method, and a typical power curve and energy usage pattern of a single user within the preset period are obtained based on the clustering quality evaluation result.

4. The method according to claim 2, characterized in that: Determining the adjustable capacity limit of the single type of distributed energy resource based on the response potential factor includes: Determining a response potential factor of the single type of distributed energy resource based on the energy usage law and energy usage mode of the single type of distributed energy resource; Obtaining a first response potential of the single type of distributed energy resource based on the response potential factor and active power of the single type of distributed energy resource; determining a second response potential of the single type of distributed energy resource based on a difference between peak and valley power of the single type of distributed energy resource; A theoretical response potential of the single type of distributed energy resource is determined based on the first response potential and the second response potential of the single type of distributed energy resource.

5. The method according to claim 4, characterized in that Also includes: A response potential assessment model for a single type of resource is constructed by introducing a cost incentive target, a peak-valley power difference minimization target, and preset constraints, wherein the preset constraints include a power reduction constraint, a response duration constraint, a response time constraint, and an energy balance constraint; The theoretical response potential of the single type of distributed energy resource is optimized using the response potential assessment model of the single type of resource to obtain the adjustable capacity limit of the single type of distributed energy resource.

6. The method according to claim 1, characterized in that The multi-type resource aggregation and interactive capability evaluation optimization model of the virtual power plant is constructed based on the regulation capability boundary of the virtual power plant in different scheduling scenarios, including: Based on the objective function of minimizing the total operating cost, the electric power balance constraint condition and the equipment constraint condition, a baseline prediction of the active power of the virtual power plant is performed to obtain a power curve of the virtual power plant when no demand response occurs; The power curve of the virtual power plant when no demand response occurs and the regulation capacity boundary under different scheduling scenarios are used to obtain the regulation capacity objective function, the ramp rate objective function and the response time objective function of the multi-type resource aggregation interaction capability evaluation optimization model; The operation constraints and state constraints of all types of distributed energy resources are used as constraints of the multi-type resource aggregation interaction capability evaluation optimization model, wherein the constraints of the multi-type resource aggregation interaction capability evaluation optimization model include climbing rate constraints, power output constraints, capacity state constraints, capacity limit constraints, response power constraints and response duration constraints.

7. The method according to claim 6, characterized in that The multi-type resource aggregation interaction capability evaluation optimization model is solved to obtain the multi-type resource aggregation interaction regulation capability of the virtual power plant, which includes: Performing a weighted operation on the objective function of the multi-type resource aggregation interaction capability evaluation optimization model to obtain a joint objective function of the multi-type resource aggregation interaction capability evaluation optimization model; Initializing the weight parameters and the required number of iterations used by the adaptive grid-based multi-objective ion swarm optimization algorithm, and generating an initial population of the adaptive grid-based multi-objective ion swarm optimization algorithm according to the joint objective function; Obtaining the fitness values ​​of particles in the initial population or the current iteration population based on the joint objective function, and determining the dominance relationship between particles in the initial population to obtain a Pareto solution set in the current iteration round; Dynamically adjusting the distance between particles of the initial population or the current iterative population using a multi-objective ion swarm optimization algorithm based on an adaptive grid to obtain an updated Pareto solution set; Iteratively performing fitness value acquisition operations, Pareto solution set acquisition operations, and population ion dynamic adjustment operations until the number of iterations is met, and then using a linear membership function to obtain the comprehensive membership of all solutions in the updated Pareto solution set; The solution with the largest comprehensive membership in the updated Pareto solution set is used as the model target solution that characterizes the multi-type resource aggregation and interactive regulation capability of the virtual power plant.

8. The method according to claim 5, characterized in that Dynamic analysis of the multi-type resource aggregation interactive regulation capability of the virtual power plant that changes dynamically over time includes: Performing data standardization on the regulation capacity, ramp rate and response time of the virtual power plant that dynamically changes over time, and obtaining information entropy of the regulation capacity, ramp rate and response time after data standardization and the redundancy of the information entropy respectively; The information entropy and the redundancy of the information entropy are used to obtain weight coefficients of the regulation capacity, the ramp rate and the response time after the data is standardized; The regulating capacity, climbing rate and response time after the data standardization are respectively averaged, and the regulating capacity, climbing rate and response time after the averaged processing are respectively calculated with the corresponding weight coefficient to obtain the dynamic analysis results of the regulating capacity, climbing rate and response time.

9. An electronic device, comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images