Dynamic aggregation method and device for incomplete load data of virtual power plant

Abstract

The invention provides a dynamic aggregation method for incomplete load data of a virtual power plant, and relates to the technical field of electric power, and the method comprises the following steps: obtaining terminal load data from a central database of the virtual power plant, and carrying out data format conversion to obtain incomplete load data in a standard format; performing grid aggregation processing on the incomplete load data by adopting a density peak clustering algorithm to obtain initialized complete load data; and using a fruit fly optimization algorithm to carry out multiple times of iterative optimization on the density peak value clustering algorithm to find an optimal aggregation coefficient in the density peak value clustering algorithm, and carrying out grid aggregation on the initialized complete load data based on an optimal aggregation center determined by the optimal aggregation coefficient to obtain complete load data. Therefore, the dynamic aggregation of the incomplete load data of the virtual power plant is reliably and comprehensively realized through the density peak value clustering algorithm after iterative optimization is performed on the density peak value clustering algorithm based on the fruit fly optimization algorithm, and a reliable basis is provided for the operation management of the virtual power plant.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method, device, electronic equipment and storage medium for dynamically aggregating incomplete load data of a virtual power plant. Background Art

[0002] During the operation of a virtual power plant, data aggregation is the basis for the virtual power plant to achieve coordinated power management of the grid operation. In addition, in the process of aggregating distributed power generation, energy storage systems, controllable loads, and electric vehicle distributed energy, the virtual power plant will store a large amount of incomplete load data, which will reduce the efficiency of coordinated control of the virtual power plant. In related technologies, since the incomplete load data of virtual power plants have large differences in structure and dimension, and the load data changes significantly dynamically and is incremental data, the aggregation effect of such data is poor. Therefore, there is an urgent need for a more reliable method for dynamic aggregation of incomplete load data of virtual power plants. Summary of the invention

[0003] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0004] To this end, the first purpose of the present invention is to propose a method for dynamically aggregating incomplete load data of a virtual power plant. By using a density peak clustering algorithm that iteratively optimizes the density peak clustering algorithm based on the fruit fly optimization algorithm, the dynamic aggregation of incomplete load data of the virtual power plant can be reliably and comprehensively realized, providing a reliable basis for the operation and management of the virtual power plant.

[0005] The second object of the present invention is to propose a device for dynamically aggregating incomplete load data of a virtual power plant.

[0006] A third objective of the present invention is to provide an electronic device.

[0007] A fourth object of the present invention is to provide a non-transitory computer-readable storage medium storing computer instructions.

[0008] A fifth object of the present invention is to provide a computer program product.

[0009] To achieve the above-mentioned purpose, a first embodiment of the present invention proposes a method for dynamically aggregating incomplete load data of a virtual power plant, the method comprising:

[0010] Obtain terminal load data from the central database of the virtual power plant, and convert the data format of each terminal load data to obtain incomplete load data in a standard format;

[0011] The density peak clustering algorithm is used to perform grid aggregation processing on the incomplete load data to obtain the initialized complete load data;

[0012] The fruit fly optimization algorithm is used to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm. Based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, the initialized complete load data is grid aggregated to obtain the complete load data of the virtual power plant.

[0013] To achieve the above-mentioned purpose, a second embodiment of the present invention proposes a device for dynamically aggregating incomplete load data of a virtual power plant, the device comprising:

[0014] An acquisition module is used to acquire terminal load data from the central database of the virtual power plant and convert the data format of each terminal load data to obtain incomplete load data in a standard format;

[0015] The first aggregation module is used to perform grid aggregation processing on the incomplete load data by using a density peak clustering algorithm to obtain initialized complete load data;

[0016] The second aggregation module is used to use the fruit fly optimization algorithm to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, perform grid aggregation on the initialized complete load data to obtain the complete load data of the virtual power plant.

[0017] To achieve the above-mentioned purpose, the third aspect of the present invention proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in the first aspect.

[0018] In order to achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to enable the computer to execute the method described in the first aspect.

[0019] The method, device, electronic device and storage medium for dynamically aggregating incomplete load data of a virtual power plant provided by the embodiment of the present invention obtain terminal load data from the central database of the virtual power plant and perform data format conversion to obtain incomplete load data in a standard format; a density peak clustering algorithm is used to perform grid aggregation processing on the incomplete load data to obtain initialized complete load data; the density peak clustering algorithm is optimized multiple times iteratively using the fruit fly optimization algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and the initialized complete load data is grid aggregated based on the optimal aggregation center determined by the optimal aggregation coefficient to obtain complete load data. Therefore, through the density peak clustering algorithm that is iteratively optimized based on the fruit fly optimization algorithm, the dynamic aggregation of incomplete load data of the virtual power plant is reliably and comprehensively realized, providing a reliable basis for the operation and management of the virtual power plant.

[0020] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0022] Figure 1 A schematic diagram of a flow chart of a method for dynamically aggregating incomplete load data of a virtual power plant provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a hierarchical structure of a central database provided by an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of a terminal load data format conversion process provided by an embodiment of the present invention;

[0025] Figure 4 A schematic flow chart of another method for dynamically aggregating incomplete load data of a virtual power plant provided by an embodiment of the present invention;

[0026] Figure 5 A structural schematic diagram of a device for dynamically aggregating incomplete load data of a virtual power plant provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0028] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of relevant laws and regulations.

[0029] The following describes a method and device for dynamically aggregating incomplete load data of a virtual power plant according to an embodiment of the present invention with reference to the accompanying drawings.

[0030] Figure 1 A schematic flow chart of a method for dynamically aggregating incomplete load data of a virtual power plant provided in an embodiment of the present invention.

[0031] like Figure 1 As shown, the method comprises the following steps:

[0032] Step 101, obtaining terminal load data from the central database of the virtual power plant, and converting the data format of each terminal load data to obtain incomplete load data in a standard format.

[0033] In some embodiments, the centralized control center is the core support for the operation of the virtual power plant. It is equipped with a central database and centralized control functions to provide data basis and control decisions for the operation analysis and decision-making of the virtual power plant. Its main work is to collect online access resource operation data, meteorological data, power grid data, etc., and integrate resources through professional management systems. Based on the integrated data, the coordinated optimization of controllable loads is achieved. The hierarchical structure diagram of the central database is shown in the figure below. Figure 2 shown.

[0034] Specifically, the central database mainly includes power load data, terminal load data (distributed power generation system data), time and weather data (meteorological data), and some commonly used basic data. This information is monitored and managed in real time by the centralized control center. It must be processed before being saved to the corresponding application of the central database, and must be continuously updated through the integrator. The integrator is mainly responsible for acquiring power load data and performing comprehensive conversion processing to ensure that the data format is correct, so as to form data support for callable resources in the coordinated scheduling of controllable loads for virtual power plants through the data analysis and processing center.

[0035] In other embodiments, the terminal load data is an important basis for realizing coordinated management of power grid energy. During the collection process, the terminal load data is affected by factors such as the collection equipment and the collection environment, which may result in some incomplete load data in the collected terminal load data. These terminal load data have the characteristics of authenticity or damage, which affects the practicality of the terminal load data.

[0036] Therefore, before the incomplete load data of the virtual power plant is aggregated, all the stored terminal load data must first be obtained through the central database. However, there are some differences in the structure and type of each data in the stored terminal load data. Therefore, before the terminal load data is aggregated, the data format needs to be converted to form a unified standard format. The terminal load data format conversion process is as follows: Figure 3 shown.

[0037] Specifically, the terminal load data is initialized, and the terminal load data is read after initialization. When the terminal load data is an empty number, the processing process of the terminal load data is terminated; when the terminal load data is not an empty number, the terminal load data is loaded, and data points are generated in time sequence; and when it is determined whether to load the data column value of the data point; if it is determined to load the data column value, the data time is written, and it is determined whether the data column value is an empty number; if the data column value is not loaded, a worksheet is created, the data time is written, and when the data column value is not written into the worksheet, a new worksheet is established until all the data column values ​​are written into the worksheet, and the data time is written after writing into the worksheet, and finally it is determined whether it is an empty number. Thus, the terminal load data format conversion is completed by writing the data column value of the data point corresponding to the terminal load data into the worksheet.

[0038] Step 102 : Using a density peak clustering algorithm, grid aggregation is performed on the incomplete load data to obtain initialized complete load data.

[0039] In some embodiments, the density peaks clustering algorithm (DPC) mainly includes three steps: first, calculating the density function threshold of the incomplete load data in the spatial grid unit; second, selecting the initial aggregation center in the dense spatial grid unit; third, each initial aggregation center aggregates the incomplete load data. In this way, dynamic aggregation of incomplete load data can be achieved.

[0040] Step 103, use the fruit fly optimization algorithm to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, grid aggregation is performed on the initialized complete load data to obtain the complete load data of the virtual power plant.

[0041] In some embodiments, the density peak clustering algorithm is iteratively optimized multiple times using a fruit fly optimization algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and the optimal aggregation center of the density peak clustering algorithm determined based on the optimal aggregation coefficient is grid aggregated on the initialized complete load data to obtain the complete load data of the virtual power plant. An implementation method may be to set the number of fruit fly populations used to represent in the fruit fly optimization algorithm and the maximum number of iterations used for expression to T; select an experimental data set from the initialized complete load data, calculate the truncation distance between the edge of each dense space grid unit in the experimental data set and the initialized aggregation center; initialize the truncation distance and the fruit fly population position; based on the initialized truncation distance , the position of the fruit fly population, the number of fruit fly populations, execute the density peak clustering algorithm, calculate the aggregation coefficient of the density peak clustering algorithm; update the initialization truncation distance for iterative optimization to record the corresponding backup aggregation coefficient; record the optimal position and value of the initialization truncation distance, if the backup aggregation coefficient value of the new position is better than the backup aggregation coefficient value of the previous generation, then update the position of the fruit fly population, otherwise keep the truncation distance position unchanged; until the maximum number of iterations is T, select the optimal truncation distance, and determine the optimal aggregation coefficient based on the optimal truncation distance, and use the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient to grid aggregate the initialized complete load data to obtain the complete load data of the virtual power plant. In this way, the aggregation effect of the density peak clustering algorithm is improved.

[0042] Specifically, the fruit fly optimization algorithm is combined with the DPC algorithm according to the fitness function, and the fruit fly optimization algorithm

[30] is used for multiple iterations to find the optimal fitness function D in the DPC algorithm. out (the cutoff distance between the edge of the grid data and the aggregation center), overcoming its sensitivity to parameter selection. The fruit fly optimization algorithm uses the silhouette coefficient Sil as the fitness function D out , and its calculation formula is:

[0043]

[0044] Where: a represents the average distance to other samples (incomplete load data) in the same category (spatial grid cell); b represents the average distance to the nearest samples in different classes. The value of the silhouette coefficient is between [-1,1]. The larger the value, the better the aggregation quality.

[0045] The detailed iterative optimization steps are as follows:

[0046] Step 1: Set the number of fruit fly populations to be represented and the maximum number of iterations to be between [31-32].

[0047] Step 2: Select the experimental data set and calculate the Euclidean distance matrix D of the experimental data setout (cutoff distance), and determine its value range as [D min ,D max ].

[0048] Step 3: Initialize D out The results and the fruit fly population position of the fruit fly optimization algorithm.

[0049] Step 4: Execute the DPC algorithm, according to D out Calculate the value of Sil.

[0050] Step 5: Update D out And calculate the corresponding Sil value.

[0051] Step 6: Record D out The optimal position and value of the new position, if the Sil value of the new position is better than the Sil value of the previous generation, the position of the fruit fly is updated, otherwise the original position remains unchanged.

[0052] Step 7: Determine whether the iteration termination condition (30 optimization iterations) is met. If the condition is met, stop the iteration and proceed to the next step; if the condition is not met, return to step 4 to continue iterative optimization.

[0053] Step 8: Set the optimal cutoff distance D out , and substitute it into the DPC algorithm to complete the grid aggregation of the initialized complete load data, and output the optimized complete load data.

[0054] The method for dynamically aggregating incomplete load data of a virtual power plant in an embodiment of the present invention obtains terminal load data from the central database of the virtual power plant and performs data format conversion to obtain incomplete load data in a standard format; a density peak clustering algorithm is used to perform grid aggregation processing on the incomplete load data to obtain initialized complete load data; the density peak clustering algorithm is optimized multiple times iteratively using the fruit fly optimization algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and the initialized complete load data is grid aggregated based on the optimal aggregation center determined by the optimal aggregation coefficient to obtain complete load data. Therefore, by using the density peak clustering algorithm after iterative optimization of the density peak clustering algorithm based on the fruit fly optimization algorithm, the dynamic aggregation of incomplete load data of the virtual power plant can be reliably and comprehensively realized, providing a reliable basis for the operation and management of the virtual power plant.

[0055] In order to clearly explain the above embodiment, Figure 4 A schematic flow chart of another method for dynamically aggregating incomplete load data of a virtual power plant provided in an embodiment of the present invention.

[0056] Step 401, obtain terminal load data from the central database of the virtual power plant, and convert the data format of each terminal load data to obtain incomplete load data in a standard format.

[0057] It should be noted that the specific implementation of step 401 can be found in the description of the relevant embodiments of the present invention, and will not be repeated here.

[0058] Step 402 : using a density peak clustering algorithm, the incomplete load data is divided into a plurality of spatial grid units, and a density function threshold for aggregating the incomplete load data in each spatial grid unit is calculated.

[0059] In some embodiments, a density peak clustering algorithm is used to divide the incomplete load data into multiple spatial grid units, and a density function threshold for aggregating the incomplete load data in each spatial grid unit is calculated. One implementation method can be to use a density peak clustering algorithm to divide the incomplete load data into multiple spatial grid units, and calculate the density function threshold for aggregating the incomplete load data in each spatial grid unit based on the number of spatial grid units, the flow coefficient, and the initial movement time and final movement time of the incomplete load data when entering the spatial grid unit.

[0060] Specifically, the incomplete load data of the virtual power plant is divided into multiple spatial grid units, each of which contains a multidimensional space s. By changing the dimension n, the incomplete load data set s={s1,s2,...,s n}Evenly divided into dn segments.

[0061] In the multidimensional data space, it is divided and processed into If m is used to represent any spatial grid unit, then all incomplete load data in the unit are represented by d = [d1, d2, ..., d n ]∈m, the constraint is

[0062] If the total number of spatial grid cells in the entire virtual power plant is represented by l, then the calculation formula is:

[0063]

[0064] The density function threshold D(m,t d ) and its calculation formula is:

[0065] D(m,t d )=λ t lD(m,t b )+1 (3)

[0066] Among them, td represents the initial moving time; t represents the flow coefficient; t b Represents the last movement time.

[0067] If the data is at the initial time t a Entering the spatial grid cell, the density function threshold for aggregating incomplete load data will be related to the change in time and is defined as:

[0068]

[0069] Where x represents incomplete load data.

[0070] If D(m,t) is used to represent the density function threshold of m-aggregated incomplete load data, at different times t and τ, the density function thresholds of the data in the following two cases are also different:

[0071]

[0072] If t a and t b Represent the initial and final moving time of m respectively, then the calculation formula of the density threshold function is:

[0073]

[0074] In addition, according to formula (6), it is possible to determine whether a spatial grid cell is a noise spatial grid cell, and determine the set D according to the time length. min The minimum value of D can identify new incomplete load data and optimize itself over time. After eliminating time, the density function threshold can be determined. When incomplete load data is received, D min (t a ,t b ) will immediately drop to the minimum, so using D min (t b ,t a ) can effectively analyze the type of spatial grid cells.

[0075] Step 403 : Based on the density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid unit is the largest is used as the initial aggregation center.

[0076] In some embodiments, based on a density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and the alternative aggregation center when the displacement of incomplete load data in each dense spatial grid unit is the largest is used as an initial aggregation center. An implementation method can be that based on a density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and the backup aggregation center of each dense spatial grid unit is selected; the displacement of incomplete load data in each dense spatial grid unit is calculated based on the distance from a point of incomplete load data in each dense spatial grid unit to the backup aggregation center, and the truncation distance between the edge of each dense spatial grid unit and the backup aggregation center; the alternative aggregation center when the displacement of incomplete load data in each dense spatial grid unit is the largest is used as the initial aggregation center.

[0077] Specifically, the purpose of selecting the initial aggregation center is to determine the distance from a point of incomplete load data to the initial aggregation center, obtain the optimal aggregation coefficient, and lay the foundation for grid aggregation. The selection of the aggregation center is related to the aggregation coefficient K. The aggregation center can be randomly selected using its average value, but the result is easily affected, resulting in a fluctuation range. If the selection is based on distance, there will be noise interference of different decibels, and the aggregation effect is not obvious. Therefore, when selecting the initial aggregation center, it is necessary to accurately determine the distance from a point of incomplete load data to the initial aggregation center to solve the distribution density problem. The dense spatial grid unit selected from multiple spatial grid units is used as the initial aggregation center with the largest displacement of incomplete load data in the dense spatial grid unit to minimize noise interference and improve the operation efficiency of the algorithm. Assuming a dense spatial grid unit as the center of the circle, the maximum distance r between incomplete load data as the radius, the circular area contained is called the r neighborhood, and the size of r is usually half of the maximum distance between incomplete load data.

[0078] If the upper threshold of p is set, there are at most p objects (incomplete load data) in the region r. Such objects are called grid center objects. The optimal distance between two spatial grid cells is:

[0079]

[0080] Where i and j are two different objects, d(i,j) is the optimal distance between two incomplete load data, n is the spatial dimension, (x i1 ,x i2 ,...,x in ) and (y i1 ,y i2 ,...,y in) is the data in n-dimensional space. The algorithm calculates the total number of objects in the region based on the optimal distance between two incomplete load data. If the number exceeds a threshold p, they can be selected and moved to a dense spatial grid cell set D. Then the initial aggregation center is found in D, and the aggregation coefficient K corresponding to the initial aggregation center is determined.

[0081] In order to make the grid aggregation result better than the original value, the incomplete load data in the spatial grid cells are separated and evaluated using the difference algorithm between the incomplete load data. Assume that there is a set of incomplete load data X = {x1, x2, ..., x N}, divide it into k groups, using {C1,C2,...,C k} represents the aggregation center of each group, that is, the aggregation centers of each group are c1, c2, ..., c k ,The distance from a point of incomplete load data in a dense spatial grid cell to the spare aggregation center is defined as:

[0082]

[0083] Due to the particularity of incomplete load data, the difference in the aggregation of incomplete load data is incomplete. Therefore, the distance between the edge of each dense spatial grid unit and the backup aggregation center is defined as the cutoff distance, and its calculation formula is:

[0084]

[0085] Where: N is the total number of incomplete load data, j The backup aggregation center of the group is c i , c i is the incomplete load data within the group, d(c i , c j ) is c i and c i The distance between them is based on the above D out , D in , and obtain an effective evaluation function F(k):

[0086]

[0087] It can be seen from formula (11) that the value range of F(k) is only guaranteed to be between [-1,1]. When the F(k) function value is close to 1, the smaller the difference in incomplete load data in each dense spatial grid unit, the more obvious the grid aggregation effect; when the function value is close to -1, it means that the grid aggregation method fails. Therefore, when the value of F(k) reaches the maximum, its aggregation coefficient K is the optimal aggregation coefficient, and the initial aggregation center at this time is selected.

[0088] Step 404 , aggregate the incomplete load data according to each initial aggregation center to obtain initialized complete load data.

[0089] In some embodiments, an implementation method of aggregating incomplete load data according to each initial aggregation center to obtain initialized complete load data can be to search for the largest dense spatial grid unit among each dense spatial grid unit, and remove, add and merge data streams of the incomplete load data through the initial aggregation center of the largest dense spatial grid unit to complete the aggregation of the incomplete load data to obtain the initialized complete load data.

[0090] Specifically, in the above processing, grid aggregation processing is used to divide the finite multidimensional space into multiple spatial grid cells. These independent spatial grid cells form a grid structure, and then the incomplete load data will be aggregated on the grid. Assume that the total number of incomplete load data in the grid structure is upper bounded by γ, and the grid cells exceeding the upper bound are called dense spatial grid cells. Define multiple dense spatial grid cells as the largest set of grid cells. Assume that the measurement unit of the incomplete load data is "block" X1, X2, ..., X i To achieve a grid structure, each block can be converted in its own grid unit, and incomplete load data can be extracted by random sampling, so the purpose is to achieve the set goal, and the incomplete load data can be aggregated by itself. The result is R1, R2, ..., R n , where R represents the number of i Each piece of data after grid aggregation.

[0091] In order to ensure that each cluster does not lose its main data information, it can randomly store multiple dense space grid cells when it moves, so all data information will be carried by the dense space grid cells, and the data information will continue to be transmitted. Assuming that the data completed at time t is X1, define the dense space network cell If we start counting each dense spatial network cell from time t The density function threshold of the data in is used as the initial cluster, then the density function threshold satisfies:

[0092]

[0093] The grid aggregation idea of ​​a single data point is as above, and the core of the data flow aggregation of the entire data set R is to first accommodate all X1, and the data it carries is regularly placed in the corresponding spatial network unit, and then the density function threshold of the data is calculated, and the maximum dense spatial grid unit among the dense spatial grid units is searched out, and the incomplete load data is removed, added, and merged through the initial aggregation center of the maximum dense spatial grid unit. The data flow is completed to aggregate the incomplete load data to obtain the initialized complete load data. For the reconnected dense spatial grid unit ξ, there will be three cases: if there is no corresponding dense spatial grid unit around, then a new dense spatial grid unit will be created and merged to obtain a new cluster; if there is a corresponding dense spatial grid unit ω, then ξ is integrated into a new dense spatial grid unit ω, if its corresponding dense spatial grid unit is ω1, ω2, ..., ω k , we can convert all clusters ω1, ω2, ..., ω k The clusters are merged into a dense space grid unit, and the fused dense space grid unit ξ will be prohibited. At this time, there will be three situations: if there are empty dense space grid units in all clusters, the dense space grid unit will be cancelled; if the dense space grid units in the cluster are all related, then the dense space grid unit ξ can be removed, and other dense space grid units remain unchanged; if all dense space grid units in the cluster are arbitrarily associated, then the cluster will be split and placed in other clusters. Under the constraints of the above conditions, the incomplete load data of the dense space grid units is aggregated again; based on the above process, the aggregation of incomplete load data of the virtual power plant is realized.

[0094] Step 405, use the fruit fly optimization algorithm to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, grid aggregation is performed on the initialized complete load data to obtain the complete load data of the virtual power plant.

[0095] The method for dynamically aggregating incomplete load data of a virtual power plant of the embodiment of the present invention obtains terminal load data from the central database of the virtual power plant, and performs data format conversion to obtain incomplete load data in a standard format; the density peak clustering algorithm is used to divide the incomplete load data into multiple spatial grid units, and the density function threshold of each spatial grid unit for aggregating incomplete load data is calculated; based on the density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid unit is the largest is used as the initial aggregation center; the incomplete load data is aggregated according to each initial aggregation center to obtain the initialized complete load data; the density peak clustering algorithm is iteratively optimized multiple times using the fruit fly optimization algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and the initialized complete load data is grid aggregated based on the optimal aggregation center determined by the optimal aggregation coefficient to obtain the complete load data. Thus, the data format conversion can be effectively completed, and the aggregation of the incomplete load data of the virtual power plant can be reliably realized, which provides a reliable basis for the operation and management of the virtual power plant.

[0096] In order to implement the above-mentioned embodiment, the present invention also proposes a device for dynamically aggregating incomplete load data of a virtual power plant.

[0097] Figure 5 A structural schematic diagram of a device for dynamically aggregating incomplete load data of a virtual power plant provided in an embodiment of the present invention.

[0098] like Figure 5 As shown, the virtual power plant incomplete load data dynamic aggregation device 50 includes: an acquisition module 51, a first aggregation module 52 and a second aggregation module 53.

[0099] The acquisition module 51 is used to acquire the terminal load data from the central database of the virtual power plant and perform data format conversion on each terminal load data to obtain incomplete load data in a standard format;

[0100] The first aggregation module 52 is used to perform grid aggregation processing on the incomplete load data by using a density peak clustering algorithm to obtain initialized complete load data;

[0101] The second aggregation module 53 is used to use the fruit fly optimization algorithm to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, perform grid aggregation on the initialized complete load data to obtain the complete load data of the virtual power plant.

[0102] Furthermore, in a possible implementation of the embodiment of the present invention, the first aggregation module 52 includes:

[0103] A calculation unit, used to divide the incomplete load data into a plurality of spatial grid units by using a density peak clustering algorithm, and calculate a density function threshold value of each spatial grid unit aggregating the incomplete load data;

[0104] A selection unit is used to select a plurality of dense spatial grid cells from the spatial grid cells based on the density function threshold, and use the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid cell is the largest as the initial aggregation center;

[0105] The aggregation unit is used to aggregate the incomplete load data according to each initial aggregation center to obtain initialized complete load data.

[0106] Furthermore, in a possible implementation manner of the embodiment of the present invention, the computing unit is specifically configured to:

[0107] The density peak clustering algorithm is used to divide the incomplete load data into multiple spatial grid units, and the density function threshold of each spatial grid unit for aggregating the incomplete load data is calculated according to the number of spatial grid units, the flow coefficient, the initial movement time and the final movement time of the incomplete load data entering the spatial grid unit.

[0108] Furthermore, in a possible implementation manner of the embodiment of the present invention, the selection unit is specifically configured to:

[0109] Based on the density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and a backup aggregation center of each dense spatial grid unit is selected;

[0110] The displacement of the incomplete load data in each dense spatial grid cell is calculated based on the distance from a point of the incomplete load data in each dense spatial grid cell to the spare aggregation center and the cutoff distance between the edge of each dense spatial grid cell and the spare aggregation center;

[0111] The alternative aggregation center when the displacement of the incomplete load data in each dense spatial grid unit is the largest is used as the initial aggregation center.

[0112] Furthermore, in a possible implementation of the embodiment of the present invention, the aggregation unit is specifically used to:

[0113] The largest dense spatial grid cell among the dense spatial grid cells is searched out, and the incomplete load data is aggregated by removing, adding and merging data streams through the initial aggregation center of the largest dense spatial grid cell to obtain the initialized complete load data.

[0114] Further, in a possible implementation of the embodiment of the present invention, the second aggregation module 53 is specifically configured to:

[0115] Set the number of fruit fly populations used to represent the fruit fly optimization algorithm and the maximum number of iterations used to represent it to T;

[0116] An experimental data set is selected from the initialized full load data, and the cutoff distance between the edge of each dense spatial grid cell in the experimental data set and the initialized aggregation center is calculated;

[0117] Initialize the cutoff distance and the position of the fruit fly population;

[0118] Based on the initial cutoff distance, the position of the fruit fly population, and the number of the fruit fly population, the density peak clustering algorithm is executed to calculate the aggregation coefficient of the density peak clustering algorithm;

[0119] Update the initialization cutoff distance for iterative optimization to record the corresponding standby aggregation coefficient;

[0120] Record the optimal position and value of the initialization cutoff distance. If the backup aggregation coefficient value of the new position is better than the backup aggregation coefficient value of the previous generation, update the position of the fruit fly population, otherwise keep the truncation distance position unchanged;

[0121] Until the maximum number of iterations is T, the optimal truncation distance is selected, and the optimal aggregation coefficient is determined based on the optimal truncation distance. The optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient is used to grid aggregate the initialized complete load data to obtain the complete load data of the virtual power plant.

[0122] It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, and will not be repeated here.

[0123] The incomplete load data dynamic aggregation device of the virtual power plant of the embodiment of the present invention obtains the terminal load data from the central database of the virtual power plant, and performs data format conversion to obtain incomplete load data in a standard format; adopts the density peak clustering algorithm to perform grid aggregation processing on the incomplete load data to obtain the initialized complete load data; uses the fruit fly optimization algorithm to perform multiple iterations of optimization on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center determined by the optimal aggregation coefficient, performs grid aggregation on the initialized complete load data to obtain the complete load data. Therefore, through the density peak clustering algorithm after iterative optimization of the density peak clustering algorithm based on the fruit fly optimization algorithm, the dynamic aggregation of the incomplete load data of the virtual power plant is reliably and comprehensively realized, providing a reliable basis for the operation and management of the virtual power plant.

[0124] In order to implement the above embodiment, the present invention further provides an electronic device, including:

[0125] at least one processor; and

[0126] a memory communicatively connected to the at least one processor; wherein,

[0127] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the aforementioned method.

[0128] In order to implement the above embodiment, the present invention further proposes a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to enable the computer to execute the above method.

[0129] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0130] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0131] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0132] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0133] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0134] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0135] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules. If the integrated module is implemented in the form of software functional modules and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0136] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for dynamically aggregating incomplete load data of a virtual power plant, characterized in that: The method comprises: Obtain terminal load data from the central database of the virtual power plant, and convert the data format of each terminal load data to obtain incomplete load data in a standard format; The density peak clustering algorithm is used to perform grid aggregation processing on the incomplete load data to obtain the initialized complete load data; The fruit fly optimization algorithm is used to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm. Based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, the initialized complete load data is grid aggregated to obtain the complete load data of the virtual power plant.

2. The method according to claim 1, characterized in that The method of using a density peak clustering algorithm to perform grid aggregation processing on the incomplete load data to obtain initialized complete load data includes: Using a density peak clustering algorithm, the incomplete load data is divided into a plurality of spatial grid units, and a density function threshold for aggregating the incomplete load data in each spatial grid unit is calculated; Based on the density function threshold, multiple dense spatial grid cells are selected from the spatial grid cells, and the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid cell is the largest is used as the initial aggregation center; The incomplete load data are aggregated according to each initial aggregation center to obtain initialized complete load data.

3. The method according to claim 2, characterized in that The method of using a density peak clustering algorithm to divide the incomplete load data into a plurality of spatial grid units and calculating a density function threshold of each spatial grid unit for aggregating the incomplete load data includes: The density peak clustering algorithm is used to divide the incomplete load data into multiple spatial grid units, and the density function threshold of each spatial grid unit for aggregating the incomplete load data is calculated according to the number of spatial grid units, the flow coefficient, the initial movement time and the final movement time of the incomplete load data entering the spatial grid unit.

4. The method according to claim 2, characterized in that: The step of selecting a plurality of dense spatial grid units from the spatial grid units based on the density function threshold, and taking the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid unit is the largest as the initial aggregation center, comprises: Based on the density function threshold, multiple dense spatial grid units are selected from the spatial grid units, and a backup aggregation center of each dense spatial grid unit is selected; The displacement of the incomplete load data in each dense spatial grid cell is calculated based on the distance from a point of the incomplete load data in each dense spatial grid cell to the spare aggregation center and the cutoff distance between the edge of each dense spatial grid cell and the spare aggregation center; The alternative aggregation center when the displacement of the incomplete load data in each dense spatial grid unit is the largest is used as the initial aggregation center.

5. The method according to claim 2, characterized in that: The aggregating the incomplete load data according to each initial aggregation center to obtain initialized complete load data includes: The largest dense spatial grid cell among the dense spatial grid cells is searched out, and the incomplete load data is aggregated by removing, adding and merging data streams through the initial aggregation center of the largest dense spatial grid cell to obtain the initialized complete load data.

6. The method according to claim 2, characterized in that The fruit fly optimization algorithm is used to perform multiple iterations of optimization on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and the optimal aggregation center of the density peak clustering algorithm determined based on the optimal aggregation coefficient is used to perform grid aggregation on the initialized complete load data to obtain the complete load data of the virtual power plant, including: Set the number of fruit fly populations used to represent the fruit fly optimization algorithm and the maximum number of iterations used to represent it to T; An experimental data set is selected from the initialized full load data, and the cutoff distance between the edge of each dense spatial grid cell in the experimental data set and the initialized aggregation center is calculated; Initialize the cutoff distance and the position of the fruit fly population; Based on the initial cutoff distance, the position of the fruit fly population, and the number of the fruit fly population, the density peak clustering algorithm is executed to calculate the aggregation coefficient of the density peak clustering algorithm; Update the initialization cutoff distance for iterative optimization to record the corresponding standby aggregation coefficient; Record the optimal position and value of the initialization cutoff distance. If the backup aggregation coefficient value of the new position is better than the backup aggregation coefficient value of the previous generation, update the position of the fruit fly population, otherwise keep the truncation distance position unchanged; Until the maximum number of iterations is T, the optimal truncation distance is selected, and the optimal aggregation coefficient is determined based on the optimal truncation distance. The optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient is used to grid aggregate the initialized complete load data to obtain the complete load data of the virtual power plant.

7. A device for dynamically aggregating incomplete load data of a virtual power plant, characterized in that: The device comprises: An acquisition module is used to acquire terminal load data from the central database of the virtual power plant and convert the data format of each terminal load data to obtain incomplete load data in a standard format; The first aggregation module is used to perform grid aggregation processing on the incomplete load data by using a density peak clustering algorithm to obtain initialized complete load data; The second aggregation module is used to use the fruit fly optimization algorithm to perform multiple iterative optimizations on the density peak clustering algorithm to find the optimal aggregation coefficient in the density peak clustering algorithm, and based on the optimal aggregation center of the density peak clustering algorithm determined by the optimal aggregation coefficient, perform grid aggregation on the initialized complete load data to obtain the complete load data of the virtual power plant.

8. The device according to claim 7, characterized in that The first aggregation module includes: A calculation unit, used to divide the incomplete load data into a plurality of spatial grid units by using a density peak clustering algorithm, and calculate a density function threshold value of each spatial grid unit aggregating the incomplete load data; A selection unit is used to select a plurality of dense spatial grid cells from the spatial grid cells based on the density function threshold, and use the candidate aggregation center when the displacement of the incomplete load data in each dense spatial grid cell is the largest as the initial aggregation center; The aggregation unit is used to aggregate the incomplete load data according to each initial aggregation center to obtain initialized complete load data.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

Images