Virtual power plant optimization scheduling method, device and equipment based on multi-objective optimization

Abstract

The invention provides a virtual power plant optimization scheduling method, device and equipment based on multi-objective optimization. The method comprises the following steps: acquiring historical power consumption data of each user in a target virtual power plant; clustering the historical power consumption data to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster; based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster, determining the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at the future moment; obtaining a scheduling target value based on the power consumption demand, the power storage quantity of distributed energy storage of the target virtual power plant and the weight of each power consumption data cluster; and determining a scheduling scheme of the target virtual power plant at the future moment by taking the lowest cost of the target virtual power plant and the minimum influence of the target virtual power plant on the power grid as targets and taking the scheduling target value as a constraint. The optimization scheduling scheme can be obtained more accurately and quickly, the dependence of the target virtual power plant on the power grid is reduced, and the stability of the power grid is maintained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid dispatching, and in particular to a virtual power plant optimal dispatching method, device and equipment based on multi-objective optimization. BACKGROUND

[0002] With the global transition to a low-carbon economy, the proportion of renewable energy such as wind and solar energy is increasing. However, these energy sources have the characteristics of intermittency and volatility, which poses challenges to the stable operation of power systems. Traditional power systems are transforming into new power systems that are more flexible, intelligent and distributed. Virtual power plants (VPPs) as an emerging power resource integration mode can effectively aggregate distributed power sources, energy storage devices and controllable loads, and realize effective support for the power grid through intelligent control strategies.

[0003] As an innovative resource integration mode, virtual power plants have significant advantages in improving the flexibility of power systems and promoting the consumption of renewable energy, but there are also some challenges and shortcomings in practical application.

[0004] Virtual power plants need to integrate data from different sources, including distributed energy resources (DERs), energy storage systems, and loads, which results in a heavy workload in organizing and operating the dispatching scheme data of virtual power plants. As a result, if a dispatching scheme is needed at a certain time, the scheme needs to be reasoned and calculated very far in advance, but the obtained scheme may not be suitable for the current state of the power grid. Therefore, such a scheme will be discarded, and there is still no accurate virtual power plant dispatching scheme. The access of virtual power plants to the power grid may cause instability of the power grid and affect the stability of the power grid. SUMMARY

[0005] The embodiments of the present application provide a virtual power plant optimal dispatching method, device and equipment based on multi-objective optimization, to obtain an optimal dispatching scheme more accurately and quickly, reduce the dependence of the target virtual power plant on the power grid, and maintain the stability of the power grid.

[0006] In a first aspect, the embodiments of the present application provide a virtual power plant optimal dispatching method based on multi-objective optimization, comprising:

[0007] Obtain historical power consumption data of each user in the target virtual power plant.

[0008] Cluster the historical power consumption data to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster; determine the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at a future time based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster.

[0009] The scheduling target value is obtained based on the electricity demand, the electricity storage of the distributed energy storage of the target virtual power plant, and the weight of each electricity data cluster.

[0010] A scheduling scheme of the target virtual power plant at a future time is determined, with a minimum cost of the target virtual power plant and a minimum influence of the target virtual power plant on the power grid as the target, and a sum of the generated electricity of the target virtual power plant and the electricity obtained by the target virtual power plant from the power grid as the scheduling target value.

[0011] In a possible implementation, the user types in the target virtual power plant include industrial users, commercial users, and residential users; the scheduling target value is obtained based on the electricity demand, the electricity storage of the distributed energy storage of the target virtual power plant, and the weight of each electricity data cluster, and includes:

[0012] The electricity gap is calculated based on the electricity demand and the electricity storage of the distributed energy storage of the target virtual power plant.

[0013] The product of the electricity gap and the weight of the target electricity data cluster is calculated to obtain the minimum generated electricity; the target electricity data cluster represents an electricity data cluster corresponding to the industrial users.

[0014] The difference between the electricity gap and the minimum generated electricity is calculated to obtain the scheduling target value.

[0015] In a possible implementation, the scheduling scheme of the target virtual power plant at a future time is determined, with a minimum cost of the target virtual power plant and a minimum influence of the target virtual power plant on the power grid as the target, and a sum of the generated electricity of the target virtual power plant and the electricity obtained by the target virtual power plant from the power grid as the scheduling target value, and includes:

[0016] The multi-objective optimization process of the minimum cost of the target virtual power plant and the minimum influence of the target virtual power plant on the power grid is solved by a multi-objective optimization algorithm, and the scheduling target value is taken as a limiting condition to obtain a plurality of Pareto optimal solutions.

[0017] The target scheme is obtained based on the weight of each electricity data cluster and the plurality of Pareto optimal solutions.

[0018] The sum of the generated electricity of the target virtual power plant in the target scheme and the minimum generated electricity is taken as the generated electricity of the target virtual power plant in the scheduling scheme of the target virtual power plant at a future time.

[0019] The electricity obtained by the target virtual power plant from the power grid in the target scheme is taken as the electricity obtained by the target virtual power plant from the power grid in the scheduling scheme of the target virtual power plant at a future time.

[0020] In a possible implementation, the target scheme is obtained based on the weight of each electricity data cluster and the plurality of Pareto optimal solutions, and includes:

[0021] determine a weight corresponding to the power consumption data cluster of the industrial user and a weight corresponding to the power consumption data cluster of the residential user based on the weight of each power consumption data cluster.

[0022] calculate a ratio of the weight corresponding to the power consumption data cluster of the industrial user and the weight corresponding to the power consumption data cluster of the residential user, denoted as a first ratio.

[0023] if the first ratio is greater than a first threshold, select a scheme with the minimum power obtained from the power grid among the plurality of Pareto optimal solutions as the target scheme.

[0024] if the first ratio is less than or equal to the first threshold, select a scheme with the minimum power generation of the target virtual power plant among the plurality of Pareto optimal solutions as the target scheme.

[0025] In a possible implementation, based on the plurality of power consumption data clusters and the cluster center corresponding to each power consumption data cluster, the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at the future time are determined, including:

[0026] determine the weight of each power consumption data cluster based on the number of users in each power consumption data cluster.

[0027] calculate the power consumption demand of the target virtual power plant at the future time based on the weight of each power consumption data cluster and the cluster center corresponding to each power consumption data cluster.

[0028] In a possible implementation, the weight of each power consumption data cluster is determined based on the number of users in each power consumption data cluster, including:

[0029] for each power consumption data cluster, calculate a ratio of the number of users in the power consumption data cluster to the total number of users to obtain the weight of the power consumption data cluster.

[0030] In a possible implementation, the power consumption demand of the target virtual power plant at the future time is calculated based on the weight of each power consumption data cluster and the cluster center corresponding to each power consumption data cluster, including:

[0031] for each power consumption data cluster, calculate a product of the cluster center corresponding to the power consumption data cluster and the weight of the power consumption data cluster to obtain a first calculation result.

[0032] calculate a sum of all the first calculation results to obtain the power consumption demand of the target virtual power plant at the future time.

[0033] In a possible implementation, the historical power consumption data of each user in the target virtual power plant is obtained, including:

[0034] obtain raw data of the historical power consumption data of each user in the target virtual power plant.

[0035] The original data is subjected to outlier cleaning and missing value completion processing to obtain historical power consumption data of each user in the target virtual power plant.

[0036] The historical power consumption data is clustered to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster, including:

[0037] For any one user, the mean value of the historical power consumption data of the user is calculated, denoted as a first mean value.

[0038] The X-means clustering is used to cluster the first mean values to obtain a plurality of power consumption data clusters.

[0039] For each power consumption data cluster, the mean value of the first mean values in the power consumption data cluster is calculated, denoted as a second mean value, and the second mean value is taken as the data center of the power consumption data cluster.

[0040] In a second aspect, an embodiment of the present application provides a virtual power plant optimization scheduling device based on multi-objective optimization, including:

[0041] The first processing module is configured to obtain historical power consumption data of each user in the target virtual power plant.

[0042] The second processing module is configured to cluster the historical power consumption data to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster, and determine a weight of each power consumption data cluster and a power consumption demand of the target virtual power plant at a future time based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster.

[0043] The third processing module is configured to obtain a scheduling target value based on the power consumption demand, a power storage capacity of distributed energy storage of the target virtual power plant, and the weight of each power consumption data cluster.

[0044] The fourth processing module is configured to determine a scheduling scheme of the target virtual power plant at the future time, with the minimum cost of the target virtual power plant and the minimum impact of the target virtual power plant on the power grid as the target, and the sum of the power generation capacity of the target virtual power plant and the power obtained by the target virtual power plant from the power grid as the constraint of the scheduling target value.

[0045] In a third aspect, an embodiment of the present application provides an electronic device including a memory and a processor, the memory storing a computer program, and the processor implementing the method in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0046] In the embodiment of the present application, the historical power consumption data of each user is first acquired, and a plurality of power consumption data clusters and their corresponding cluster centers are calculated, then the relatively accurate power consumption demand is calculated according to the plurality of power consumption data clusters and their corresponding cluster centers, then the sum (dispatch target value) of the power amount that the virtual power plant needs to obtain from the power grid and the power generation amount is accurately calculated according to the power consumption demand, the power storage amount and the weight of each power consumption data cluster, finally, the multi-objective optimization is performed with the minimum target virtual power plant cost and the minimum target virtual power plant influence on the power grid as the target, and the convergence speed of the multi-objective optimization is further accelerated by limiting the solution range of the multi-objective optimization through the dispatch target value, so that the speed of the scheme calculation result can be ensured, and the timeliness of the scheme can be ensured, which can well adapt to the state of the power grid, and the dependence of the virtual power plant on the power grid can be ensured and the stability of the power grid can be maintained due to the more accurate dispatch target value. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is an implementation flowchart of the virtual power plant optimization scheduling method based on multi-objective optimization provided by the embodiment of the present application;

[0048] Figure 2 is a structural schematic diagram of the virtual power plant optimization scheduling device based on multi-objective optimization provided by the embodiment of the present application;

[0049] Figure 3 is a schematic diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0050] The embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0051] Reference is made to Figure 1 which shows the implementation flowchart of the virtual power plant optimization scheduling method based on multi-objective optimization provided by the embodiment of the present application, and is described in detail as follows:

[0052] In step 101, the historical power consumption data of each user in the target virtual power plant is acquired.

[0053] In a possible implementation manner, step 101 can include:

[0054] The original data of the historical power consumption data of each user in the target virtual power plant is acquired.

[0055] The original data is subjected to outlier cleaning and missing value completion processing to obtain the historical power consumption data of each user in the target virtual power plant.

[0056] Exemplarily, for a user, the historical power consumption data processed by the outlier cleaning and the missing value completion can be the power consumption data of the user for consecutive days, and the power consumption demand in the future can be calculated more accurately by using the power consumption data for consecutive days, thereby providing data support for stable work of the virtual power plant.

[0057] Exemplarily, assuming that the original data of the historical power consumption data of the user A is {20, 21, 230, 19, 24, X, 22}, the "230" in the historical power consumption data of the user A is too large from other data, and can be determined as an outlier, and X represents a missing value. When X is completed, the average value of 24 and 22 can be calculated by using the adjacent "24" and "22" of X, and the average value 23 is filled in the position of X, so that the historical power consumption data of the user A becomes {20, 21, 19, 24, 23, 22}.

[0058] In step 102, the historical power consumption data is clustered to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster; and based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster, the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at the future time are determined.

[0059] In a possible implementation, the historical power consumption data is clustered to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster, including:

[0060] For any one user, the mean value of the historical power consumption data of the user is calculated, denoted as a first mean value.

[0061] The first mean values are clustered by X-means clustering to obtain a plurality of power consumption data clusters.

[0062] For each power consumption data cluster, the mean value of each first mean value in the power consumption data cluster is calculated, denoted as a second mean value, and the second mean value is taken as the data center of the power consumption data cluster.

[0063] Exemplarily, the X-means clustering method can more accurately divide the users into different power consumption data clusters. Compared with the traditional clustering of a specified number, the method is more flexible, and the obtained result is more accurate. Specifically, the user types in the target virtual power plant include industrial users, commercial users and residential users. However, it does not mean that one clustering cluster can completely represent one user type during clustering, because the actual power consumption of different industries is not completely the same. Therefore, in order to more accurately obtain the power consumption demand, the X-means clustering method is selected, so that the obtained clustering cluster is more reasonable.

[0064] In a possible implementation, the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at the future time are determined based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster, and the method comprises the following steps.

[0065] The weight of each power consumption data cluster is determined based on the number of users in each power consumption data cluster.

[0066] The power consumption demand of the target virtual power plant at the future time is calculated based on the weight of each power consumption data cluster and the clustering center corresponding to each power consumption data cluster.

[0067] By way of example, the weight of each power consumption data cluster is calculated, so that the component of the power consumption demand corresponding to each power consumption data cluster can be more accurately calculated. Compared with the overall direct estimation method, the overall is divided into a plurality of components, and the power consumption demand is calculated based on the weight of each component, which is more accurate.

[0068] In a possible implementation, the weight of each power consumption data cluster is determined based on the number of users in each power consumption data cluster, and the method comprises the following steps.

[0069] For each power consumption data cluster, the ratio of the number of users in the power consumption data cluster to the total number of users is calculated to obtain the weight of the power consumption data cluster.

[0070] In a possible implementation, the power consumption demand of the target virtual power plant at the future time is calculated based on the weight of each power consumption data cluster and the clustering center corresponding to each power consumption data cluster, and the method comprises the following steps.

[0071] For each power consumption data cluster, the product of the clustering center corresponding to the power consumption data cluster and the weight of the power consumption data cluster is calculated to obtain a first calculation result.

[0072] The sum of all the first calculation results is calculated to obtain the power consumption demand of the target virtual power plant at the future time.

[0073] In step 103, the scheduling target value is obtained based on the power consumption demand, the power storage capacity of the distributed energy storage of the target virtual power plant, and the weight of each power consumption data cluster.

[0074] In a possible implementation, when all the industrial users correspond to one power consumption data cluster, the step 103 can comprise the following steps.

[0075] The power gap amount is calculated based on the power consumption demand and the power storage capacity of the distributed energy storage of the target virtual power plant.

[0076] The product of the power gap amount and the weight of the target power consumption data cluster is calculated to obtain the minimum power generation amount, wherein the target power consumption data cluster represents the power consumption data cluster corresponding to the industrial user.

[0077] The difference between the power gap and the minimum power generation is calculated to obtain the scheduling target value.

[0078] In a possible implementation, when all industrial users correspond to multiple power consumption data clusters, the step 103 can further include:

[0079] The power gap is calculated based on the power consumption demand and the power storage capacity of the distributed energy storage of the target virtual power plant.

[0080] The sum of the product of the power gap and the weight of the target power consumption data cluster is calculated to obtain the minimum power generation; wherein the target power consumption data cluster represents the power consumption data cluster corresponding to the industrial user.

[0081] The difference between the power gap and the minimum power generation is calculated to obtain the scheduling target value.

[0082] Illustratively, in order to reduce the dependence of the target virtual power plant on the power grid and maintain the stability of the power grid, the minimum power generation is additionally set, because the existence of the minimum power generation fully considers the existence of the target power consumption data cluster, adaptively reduces the possibility of the industrial user directly accessing the power grid, and thus the stability of the power grid can be ensured.

[0083] The step 104 determines the scheduling scheme of the target virtual power plant at the future time, with the minimum target virtual power plant cost and the minimum target virtual power plant impact on the power grid as the target, and the sum of the power generation of the target virtual power plant and the power obtained by the target virtual power plant accessing the power grid as the scheduling target value as the constraint.

[0084] In a possible implementation, the step 104 can include:

[0085] The multi-objective optimization process of the minimum target virtual power plant cost and the minimum target virtual power plant impact on the power grid is solved by a multi-objective optimization algorithm, and the scheduling target value is taken as a limiting condition to obtain multiple Pareto optimal solutions. The multi-objective optimization algorithm can be an NSGA-II (Non-dominated Sorting Genetic Algorithm II) algorithm or an MOEA / D (Multi-objective Evolutionary Algorithm based on Decomposition) algorithm.

[0086] The target scheme is obtained based on the weight of each power consumption data cluster and the multiple Pareto optimal solutions.

[0087] The sum of the power generation of the target virtual power plant in the target scheme and the minimum power generation is taken as the power generation of the target virtual power plant in the scheduling scheme of the target virtual power plant at the future time.

[0088] The power obtained by the target virtual power plant from the power grid in the target scheme is taken as the power obtained by the target virtual power plant from the power grid in the scheduling scheme of the future time.

[0089] In a possible implementation, the target scheme is obtained based on the weight of each power consumption data cluster and the plurality of Pareto optimal solutions, and includes:

[0090] Based on the weight of each power consumption data cluster, the weight corresponding to the power consumption data cluster of the industrial user and the weight corresponding to the power consumption data cluster of the residential user are determined.

[0091] The ratio of the weight corresponding to the power consumption data cluster of the industrial user to the weight corresponding to the power consumption data cluster of the residential user is calculated, and is recorded as a first ratio.

[0092] If the first ratio is greater than a first threshold, a scheme with the minimum power obtained from the power grid in the plurality of Pareto optimal solutions is selected as the target scheme.

[0093] If the first ratio is less than or equal to the first threshold, a scheme with the minimum power generated by the target virtual power plant in the plurality of Pareto optimal solutions is selected as the target scheme.

[0094] For example, in order to ensure that the obtained solution (target scheme) can be more suitable for the actual situation of the target virtual power plant (the proportion of industrial users, commercial users and residential users), the first ratio is calculated. When the weight corresponding to the power consumption data cluster of the industrial user is greater than the weight corresponding to the power consumption data cluster of the residential user, it should be avoided that the virtual power plant directly obtains power from the power grid. Although a series of settings for stabilizing the power grid have been made, there may still be occasional occurrences. Therefore, when the proportion of industrial users is large (relative to the proportion of residential users), a scheme with the minimum power obtained from the power grid in the plurality of Pareto optimal solutions is selected as the target scheme. Conversely, when the proportion of industrial users is small (relative to the proportion of residential users), a scheme with the minimum power generated by the target virtual power plant in the plurality of Pareto optimal solutions is selected as the target scheme.

[0095] The virtual power plant optimization scheduling method based on multi-objective optimization provided by the embodiment of the application first obtains the historical power consumption data of each user, and calculates a plurality of power consumption data clusters and corresponding clustering centers, then calculates the relatively accurate power consumption demand according to the plurality of power consumption data clusters and the corresponding clustering centers, and then accurately calculates the sum of the power amount that the virtual power plant needs to obtain from the power grid and the power generation amount (scheduling target value) according to the power consumption demand, the power storage amount of the distributed energy storage of the target virtual power plant and the weight of each power consumption data cluster, and finally performs multi-objective optimization with the lowest target virtual power plant cost and the smallest influence of the target virtual power plant on the power grid as the target, and further limits the solution range of the multi-objective optimization through the scheduling target value, which can accelerate the convergence speed of the multi-objective optimization, thus ensuring the speed of the calculation result of the scheme, and further ensuring the timeliness of the scheme, which can well adapt to the state of the power grid, and in addition, due to the more accurate scheduling target value, the dependence of the virtual power plant on the power grid can be ensured, and the stability of the power grid can be maintained.

[0096] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0097] The following is a device embodiment of the application, and for details not described in detail, reference can be made to the corresponding method embodiments described above.

[0098] Figure 2 The structure schematic diagram of the virtual power plant optimization scheduling device based on multi-objective optimization provided by the embodiment of the application is shown, only the parts related to the embodiment of the application are shown for the convenience of description, and the details are as follows:

[0099] As shown in Figure 2 The virtual power plant optimization scheduling device based on multi-objective optimization comprises:

[0100] The first processing module 201 is configured to obtain the historical power consumption data of each user in the target virtual power plant.

[0101] The second processing module 202 is configured to cluster the historical power consumption data to obtain a plurality of power consumption data clusters and a clustering center corresponding to each power consumption data cluster, and determine the weight of each power consumption data cluster and the power consumption demand of the target virtual power plant at a future time based on the plurality of power consumption data clusters and the clustering center corresponding to each power consumption data cluster.

[0102] The third processing module 203 is configured to obtain a scheduling target value based on the power consumption demand, the power storage amount of the distributed energy storage of the target virtual power plant and the weight of each power consumption data cluster.

[0103] The fourth processing module 204 is configured to determine a scheduling scheme of the target virtual power plant at the future moment, with the minimum cost of the target virtual power plant and the minimum influence of the target virtual power plant on the power grid as the target, and with the sum of the power generation of the target virtual power plant and the power obtained by the target virtual power plant from the power grid as a scheduling target value.

[0104] In a possible implementation, the user types in the target virtual power plant include industrial users, commercial users, and residential users; and the third processing module 203 can be configured to:

[0105] The power gap is calculated based on the power demand and the power storage capacity of the distributed energy storage of the target virtual power plant.

[0106] The product of the power gap and the weight of the target power consumption data cluster is calculated to obtain the minimum power generation; wherein the target power consumption data cluster represents the power consumption data cluster corresponding to the industrial user.

[0107] The difference between the power gap and the minimum power generation is calculated to obtain the scheduling target value.

[0108] In a possible implementation, the fourth processing module 204 can be configured to:

[0109] The multi-objective optimization process of the minimum cost of the target virtual power plant and the minimum influence of the target virtual power plant on the power grid is solved by using a multi-objective optimization algorithm, and the scheduling target value is taken as a limiting condition to obtain a plurality of Pareto optimal solutions.

[0110] The target scheme is obtained based on the weight of each power consumption data cluster and the plurality of Pareto optimal solutions.

[0111] The sum of the power generation of the target virtual power plant in the target scheme and the minimum power generation is taken as the power generation of the target virtual power plant in the scheduling scheme of the target virtual power plant at the future moment.

[0112] The power obtained by the target virtual power plant in the target scheme from the power grid is taken as the power obtained by the target virtual power plant from the power grid in the scheduling scheme of the target virtual power plant at the future moment.

[0113] In a possible implementation, the fourth processing module 204 can be configured to:

[0114] The weight corresponding to the power consumption data cluster of the industrial user and the weight corresponding to the power consumption data cluster of the residential user are determined based on the weight of each power consumption data cluster.

[0115] The ratio of the weight corresponding to the power consumption data cluster of the industrial user to the weight corresponding to the power consumption data cluster of the residential user is calculated and recorded as a first ratio.

[0116] If the first ratio is greater than the first threshold, a scheme with the least power obtained from the grid among the plurality of Pareto optimal solutions is selected as the target scheme.

[0117] If the first ratio is less than or equal to the first threshold, a scheme with the least power generated by the target virtual power plant among the plurality of Pareto optimal solutions is selected as the target scheme.

[0118] In a possible implementation, the second processing module 202 can be configured to:

[0119] Determine a weight of each power consumption data cluster based on a number of users in the power consumption data cluster.

[0120] Calculate the power consumption demand of the target virtual power plant at the future moment based on the weight of each power consumption data cluster and the clustering center corresponding to each power consumption data cluster.

[0121] In a possible implementation, the weight of each power consumption data cluster is determined based on the number of users in the power consumption data cluster, and the weight of each power consumption data cluster comprises:

[0122] For each power consumption data cluster, a ratio of the number of users in the power consumption data cluster to the total number of users is calculated to obtain the weight of the power consumption data cluster.

[0123] In a possible implementation, the second processing module 202 can be configured to:

[0124] For each power consumption data cluster, a product of the clustering center corresponding to the power consumption data cluster and the weight of the power consumption data cluster is calculated to obtain a first calculation result.

[0125] A sum of all the first calculation results is calculated to obtain the power consumption demand of the target virtual power plant at the future moment.

[0126] In a possible implementation, the first processing module 201 can be configured to:

[0127] Obtain original data of historical power consumption data of each user in the target virtual power plant.

[0128] Perform outlier cleaning and missing value completion processing on the original data to obtain the historical power consumption data of each user in the target virtual power plant.

[0129] In a possible implementation, the second processing module 202 can be configured to:

[0130] For any one user, a mean value of the historical power consumption data of the user is calculated, denoted as a first mean value.

[0131] Each first mean value is clustered by X-means clustering to obtain a plurality of power consumption data clusters.

[0132] For each power consumption data cluster, a mean of the first means in the power consumption data cluster is calculated, denoted as a second mean, and the second mean is taken as the data center of the power consumption data cluster.

[0133] Figure 3 is a schematic diagram of an electronic device provided by an embodiment of the present application. As shown in the figure, the electronic device 3 of this embodiment comprises a processor 30 and a memory 31. The memory 31 stores a computer program 32. The processor 30 implements the steps in each of the above method embodiments when executing the computer program 32. Alternatively, the processor 30 implements the functions of each module / unit in each of the above device embodiments when executing the computer program 32. Figure 3

[0134] By way of example, the computer program 32 can be divided into one or more modules / units, which are stored in the memory 31 and executed by the processor 30 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 32 in the electronic device 3.

[0135] The electronic device 3 can include, but is not limited to, the processor 30 and the memory 31. Those skilled in the art can understand that Figure 3 The electronic device 3 is merely an example and does not constitute a limitation on the electronic device 3, and can include more or fewer components than those shown, or combine certain components, or different components, for example, the electronic device 3 can also include an input / output device, a network access device, a bus, etc.

[0136] For the convenience and brevity of description, only the above-mentioned division of functional modules / units is exemplified, and in actual application, the above-mentioned functions can be completed by different functional modules / units according to needs. The above-mentioned modules / units can be realized in the form of hardware, software, or a combination of hardware and software.

[0137] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments. If there is no special description and no logical conflict, the terms and / or descriptions of different embodiments are consistent and can be mutually referenced, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0138] ​The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A virtual power plant optimization scheduling method based on multi-objective optimization, characterized in that: include: Obtain historical electricity consumption data for each user in the target virtual power plant; Clustering the historical electricity consumption data to obtain multiple electricity consumption data clusters and a cluster center corresponding to each electricity consumption data cluster; Determining, based on the multiple power consumption data clusters and the cluster center corresponding to each power consumption data cluster, a weight of each power consumption data cluster and a power demand of the target virtual power plant at a future time; Obtaining a scheduling target value based on the electricity demand, the amount of distributed energy storage of the target virtual power plant, and the weight of each electricity data cluster; With the goal of minimizing the cost of the target virtual power plant and the impact of the target virtual power plant on the power grid, and with the sum of the power generation of the target virtual power plant and the power obtained by the target virtual power plant from connecting to the power grid as the scheduling target value as constraints, the scheduling plan for the target virtual power plant at future times is determined.

2. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 1 is characterized in that: The user types in the target virtual power plant include: industrial users, commercial users, and residential users; the scheduling target value is obtained based on the power demand, the storage capacity of the distributed energy storage of the target virtual power plant, and the weight of each power data cluster, including: Calculating the power gap based on the power demand and the distributed energy storage capacity of the target virtual power plant; Calculating the product of the power gap and the weight of the target power consumption data cluster to obtain the minimum power generation; wherein the target power consumption data cluster represents the power consumption data cluster corresponding to industrial users; The difference between the power gap and the minimum power generation is calculated to obtain a scheduling target value.

3. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 2 is characterized in that: The method determines the scheduling plan for the target virtual power plant at a future time, with the lowest cost of the target virtual power plant and the smallest impact of the target virtual power plant on the power grid as the goals, and with the sum of the power generation of the target virtual power plant and the power obtained by the target virtual power plant when connected to the power grid as the scheduling target value as a constraint, including: Solving a multi-objective optimization process of minimizing the cost of a target virtual power plant and minimizing the impact of the target virtual power plant on the power grid through a multi-objective optimization algorithm, and taking the scheduling target value as a constraint condition to obtain multiple Pareto optimal solutions; Obtaining a target solution based on the weight of each electricity consumption data cluster and the multiple Pareto optimal solutions; The sum of the power generation of the target virtual power plant in the target plan and the minimum power generation is used as the power generation of the target virtual power plant in the scheduling plan of the target virtual power plant at a future time; The amount of electricity obtained by the target virtual power plant in the target plan when connected to the power grid is used as the amount of electricity obtained by the target virtual power plant in the scheduling plan of the target virtual power plant at a future moment when connected to the power grid.

4. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 3 is characterized in that: The obtaining of a target solution based on the weight of each electricity consumption data cluster and the multiple Pareto optimal solutions includes: Determining, based on the weight of each electricity consumption data cluster, a weight corresponding to the electricity consumption data cluster of industrial users and a weight corresponding to the electricity consumption data cluster of residential users; Calculate the ratio of the weight corresponding to the electricity consumption data cluster of industrial users to the weight corresponding to the electricity consumption data cluster of residential users, and record it as a first ratio; If the first ratio is greater than a first threshold, selecting a solution with the minimum amount of electricity obtained by connecting to the power grid among multiple Pareto optimal solutions as the target solution; If the first ratio is less than or equal to the first threshold, the solution with the minimum power generation of the target virtual power plant among multiple Pareto optimal solutions is selected as the target solution.

5. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 1 is characterized in that: The determining, based on the multiple power consumption data clusters and the cluster center corresponding to each power consumption data cluster, the weight of each power consumption data cluster and the power demand of the target virtual power plant at a future moment includes: Determining a weight for each electricity consumption data cluster based on the number of users in each electricity consumption data cluster; Based on the weight of each electricity consumption data cluster and the cluster center corresponding to each electricity consumption data cluster, the electricity demand of the target virtual power plant at a future moment is calculated.

6. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 5 is characterized in that: The determining the weight of each electricity consumption data cluster based on the number of users in each electricity consumption data cluster includes: For each electricity consumption data cluster, the ratio of the number of users in the electricity consumption data cluster to the total number of users is calculated to obtain the weight of the electricity consumption data cluster.

7. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 5 is characterized in that: The calculating of the power demand of the target virtual power plant at a future time based on the weight of each power consumption data cluster and the cluster center corresponding to each power consumption data cluster includes: For each electricity consumption data cluster, calculating the product of the cluster center corresponding to the electricity consumption data cluster and the weight of the electricity consumption data cluster to obtain a first calculation result; Calculate the sum of all the first calculation results to obtain the electricity demand of the target virtual power plant at a future time.

8. The virtual power plant optimization scheduling method based on multi-objective optimization according to claim 1 is characterized in that: The obtaining of historical electricity consumption data of each user in the target virtual power plant includes: Obtain the original data of historical electricity consumption data of each user in the target virtual power plant; Perform outlier cleaning and missing value filling processing on the original data to obtain historical electricity consumption data of each user in the target virtual power plant; The clustering of the historical electricity consumption data to obtain a plurality of electricity consumption data clusters and a cluster center corresponding to each electricity consumption data cluster includes: For any user, calculate the mean of the user's historical electricity consumption data, which is recorded as the first mean; Clustering each of the first means by X-means clustering to obtain multiple electricity consumption data clusters; For each power consumption data cluster, the mean of the first mean values ​​in the power consumption data cluster is calculated and recorded as a second mean value, and the second mean value is used as the data center of the power consumption data cluster.

9. A virtual power plant optimization scheduling device based on multi-objective optimization, characterized in that: include: The first processing module is used to obtain historical electricity consumption data of each user in the target virtual power plant; A second processing module is used to cluster the historical electricity consumption data to obtain multiple electricity consumption data clusters and a cluster center corresponding to each electricity consumption data cluster; Determining, based on the multiple power consumption data clusters and the cluster center corresponding to each power consumption data cluster, a weight of each power consumption data cluster and a power demand of the target virtual power plant at a future time; A third processing module is configured to obtain a scheduling target value based on the power demand, the amount of distributed energy storage of the target virtual power plant, and the weight of each power data cluster; The fourth processing module is used to determine the scheduling plan of the target virtual power plant at a future time with the goal of minimizing the cost of the target virtual power plant and the impact of the target virtual power plant on the power grid, and with the sum of the power generation of the target virtual power plant and the power obtained by the target virtual power plant from connecting to the power grid as the scheduling target value as a constraint.

10. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images