Virtual power plant resource partitioning method based on multi-layer perceptron

Abstract

The invention provides a virtual power plant resource partitioning method based on a multi-layer perceptron. The method comprises the following steps: acquiring geographical location information of each resource in a virtual power plant; adjusting a density-based clustering algorithm according to the geographical location information of each resource in the virtual power plant, and performing clustering analysis on the geographical location information of each resource by using the adjusted density-based clustering algorithm to obtain a plurality of first clustering results; counting the number of geographical location information contained in each first clustering result, adjusting the first clustering results according to the number to obtain a plurality of second clustering results, and taking a range formed by each second clustering result as a partition; and inputting the geographical location information of each resource into a pre-trained multi-layer perceptron to obtain an importance degree score of each resource. According to the method provided by the invention, each energy resource can be partitioned and effectively regulated and controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a virtual power plant resource partitioning method based on a multi-layer perceptron. Background Art

[0002] Virtual Power Plants (VPPS) was first proposed in 1997 and attracted the attention of European and North American countries in the early 21st century. Now, many mature demonstration projects have been formed, playing an important role in absorbing distributed energy and stabilizing the power market. For my country, VPPS, as an emerging technology, is still in its infancy.

[0003] A virtual power plant refers to a power coordination management system that uses advanced communication technology and software architecture to achieve the aggregation and coordinated optimization of various geographically dispersed distributed energy sources, so as to participate in the power market and power grid operation as a special power plant. A virtual power plant aggregates controllable units, distributed energy, energy storage equipment, loads, electric vehicles, communication equipment, etc. on the energy cloud, and realizes its functions through unified coordination through technical means. Among them, virtual power plant zoning occupies an important position in the work of virtual power plants, so how to effectively zoning plays an important role in the operation and regulation of virtual power plants. Summary of the invention

[0004] The purpose of the present invention is to provide a virtual power plant resource partitioning method based on a multi-layer perceptron to improve the above-mentioned problem.

[0005] In order to achieve the above objectives, the present application provides the following technical solutions:

[0006] A virtual power plant resource partitioning method based on a multi-layer perceptron, comprising:

[0007] Obtain the geographic location information of each resource in the virtual power plant;

[0008] Adjusting the density-based clustering algorithm according to the geographic location information of each resource in the virtual power plant, and performing cluster analysis on the geographic location information of each resource using the adjusted density-based clustering algorithm to obtain a plurality of first clustering results;

[0009] Counting the number of geographical location information included in each first clustering result, adjusting the first clustering result according to the number, obtaining a plurality of second clustering results, each of which includes the same number of geographical location information, and treating the range formed by each second clustering result as a partition;

[0010] The geographical location information of each resource is input into the pre-trained multi-layer perceptron to obtain the importance score of each resource, and the total importance score corresponding to each partition is sent to the staff to help the staff manage each partition.

[0011] Furthermore, the density-based clustering algorithm is adjusted according to the geographical location information of each resource in the virtual power plant, including:

[0012] The geographical location information of each resource in the virtual power plant is collected to obtain a data set G, and a neighborhood radius list is generated according to the data set G;

[0013] A neighborhood minimum point list is generated according to the neighborhood radius list; and parameters of a density-based clustering algorithm are adjusted according to the neighborhood radius list and the neighborhood minimum point list to obtain an adjusted clustering algorithm.

[0014] Furthermore, a neighborhood radius list is generated according to the data set G, including:

[0015] Calculate the distance distribution matrix of each geographic location information in the data set;

[0016] The data in each row of the distance distribution matrix are sorted from small to large. After sorting, the quartile method is used to remove isolated points from each column of the distance distribution matrix, and then the expectation of the retained data points is calculated to obtain the first value; all the calculated first values ​​are saved in the neighborhood radius list.

[0017] Furthermore, the distance distribution matrix of each geographic location information in the data set is calculated, including:

[0018] The distance distribution matrix of each geographic location information in the data set is calculated according to formula (1), and the formula (1) is:

[0019] G n×n ={Point_Dist(i,j)1≤i≤n,1≤j≤n} (1)

[0020] In formula (1), G n×n is an n×n symmetric matrix; n is the length of the data set G; Point_Dist(i,j) is the distance between the i-th geographic location information and the j-th geographic location information in the data set G.

[0021] Further, a neighborhood minimum point list is generated according to the neighborhood radius list; and a density-based clustering algorithm is parameter-adjusted according to the neighborhood radius list and the neighborhood minimum point list to obtain an adjusted clustering algorithm, including:

[0022] For all the sample points contained in the neighborhood corresponding to each neighborhood radius in the neighborhood radius list, the mathematical expectation value is used as a parameter in the neighborhood minimum point list to obtain the neighborhood minimum point list;

[0023] A value is arbitrarily selected from the neighborhood radius list and the neighborhood minimum point number list as a parameter in the density-based clustering algorithm. After the selection, a clustering analysis is performed using the density-based clustering algorithm to obtain multiple third clustering results, and an adjusted clustering algorithm is obtained based on the multiple third clustering results.

[0024] Furthermore, the adjusted clustering algorithm is obtained based on the multiple third clustering results, including:

[0025] The silhouette coefficient method is used to calculate the silhouette coefficients corresponding to the multiple third clustering results; the silhouette coefficient is analyzed, wherein if the silhouette coefficient is greater than a preset threshold, the currently selected neighborhood radius and the minimum number of neighborhood points are used as final parameters of the density-based clustering algorithm, otherwise, a value is randomly selected from the neighborhood radius list and the minimum number of neighborhood points list as a parameter in the density-based clustering algorithm, and clustering and silhouette coefficient calculation are performed again until the silhouette coefficient is greater than the preset threshold.

[0026] Furthermore, the number of geographical location information included in each first clustering result is counted, and the first clustering result is adjusted according to the number to obtain multiple second clustering results, including:

[0027] The number of geographic location information included in each first clustering result is counted, and if the number is not equal to a preset value, the geographic location information included in each first clustering result is adjusted, so that the number of geographic location information included in each first clustering result is equal to the preset value after adjustment.

[0028] Further, the geographical location information included in each first clustering result is adjusted so that the number of geographical location information included in each first clustering result is equal to a preset value after the adjustment, including:

[0029] The first clustering result whose number of geographic location information is equal to the preset value is recorded as a cluster that does not need to be adjusted; the first clustering result whose number of geographic location information is not equal to the preset value is recorded as a cluster that needs to be adjusted; the distance from each cluster that needs to be adjusted to each cluster that does not need to be adjusted is calculated, and the corresponding distances of each cluster that needs to be adjusted are added to obtain a first value, and traversal is started from the cluster that needs to be adjusted with the largest first value, and adjustments are made so that the number of geographic location information contained in each first clustering result is equal to the preset value.

[0030] Further, traversing from the cluster to be adjusted with the largest first value and making adjustments, after the adjustments, the number of geographical location information included in each first clustering result is equal to a preset value, including:

[0031] The traversal is started from the cluster requiring adjustment with the largest first value. When the number of geographic location information contained in the cluster requiring adjustment with the largest first value is less than the preset value, the geographic location information closest to the centroid of the cluster requiring adjustment with the largest first value is added to the cluster requiring adjustment with the largest first value, so that the number of geographic location information contained in the cluster requiring adjustment with the largest first value is equal to the preset value. When the number of geographic location information contained in the cluster requiring adjustment with the largest first value is greater than the preset value, the geographic location information closest to the centroid of all geographic location information in the cluster requiring adjustment with the largest first value is removed from the cluster requiring adjustment with the largest first value in turn, until the number of geographic location information contained in the cluster requiring adjustment with the largest first value is equal to the preset value. Meanwhile, during the adjustment process, when the number of geographic location information contained is equal to the preset value, the first clustering result is recorded as a cluster that does not require adjustment.

[0032] Furthermore, the range formed by each second clustering result is taken as a partition, including:

[0033] The geographical range formed on the map by all the geographical location information included in each second clustering result is regarded as a partition.

[0034] In a second aspect, an embodiment of the present application provides a virtual power plant resource partitioning device based on a multi-layer perceptron, comprising:

[0035] An acquisition module is used to obtain the geographical location information of each resource in the virtual power plant;

[0036] A clustering module, used to adjust the density-based clustering algorithm according to the geographical location information of each resource in the virtual power plant, and use the adjusted density-based clustering algorithm to perform cluster analysis on the geographical location information of each resource to obtain multiple first clustering results;

[0037] an adjustment module, configured to count the number of geographical location information included in each first clustering result, adjust the first clustering result according to the number, obtain multiple second clustering results, each of which includes the same number of geographical location information, and use the range formed by each second clustering result as a partition;

[0038] The management module is used to input the geographical location information of each resource into a pre-trained multi-layer perceptron to obtain the importance score of each resource, and send the total importance score corresponding to each partition to the staff to help the staff manage each partition.

[0039] In a third aspect, an embodiment of the present application provides a debris flow risk monitoring device, the device comprising a memory and a processor. The memory is used to store a computer program; the processor is used to implement the steps of the virtual power plant resource partitioning method based on a multilayer perceptron when executing the computer program.

[0040] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned virtual power plant resource partitioning method based on a multi-layer perceptron are implemented.

[0041] The beneficial effects of the present invention are:

[0042] In the present invention, the geographical location of each energy resource is firstly obtained, and a cluster analysis is performed from the perspective of geographical location to form a plurality of clustering results. Meanwhile, during clustering, the traditional density-based clustering algorithm is improved, and the clustering effect of the improved clustering algorithm is more superior. After clustering, each clustering result is dynamically adjusted in consideration of the uniformity of resources in each partition, and the resources in each partition can be balanced through the adjustment method. Finally, the importance of each partition resource is scored, and this method can help the staff understand the importance of each partition, which is more conducive to management and regulation.

[0043] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 It is a schematic diagram of a virtual power plant resource partitioning method based on a multi-layer perceptron in an embodiment of the present invention;

[0046] Figure 2 It is a schematic diagram of the structure of a virtual power plant resource partitioning device based on a multi-layer perceptron in an embodiment of the present invention;

[0047] Figure 3 It is a schematic diagram of the structure of a virtual power plant resource partitioning device based on a multi-layer perceptron described in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0049] It should be noted that similar reference numerals or letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment provides a virtual power plant resource partitioning method based on a multi-layer perceptron, and the method includes step S1, step S2, step S3 and step S4.

[0052] Step S1, obtaining geographical location information of each resource in the virtual power plant;

[0053] In this step, each resource can be understood as an energy resource, i.e., an energy supply object;

[0054] Step S2: adjusting the density-based clustering algorithm according to the geographical location information of each resource in the virtual power plant, and performing cluster analysis on the geographical location information of each resource using the adjusted density-based clustering algorithm to obtain a plurality of first clustering results;

[0055] In this step, the density-based clustering algorithm is the DBSCAN clustering algorithm, and the specific implementation steps of this step include step S21 and step S22;

[0056] Step S21, collecting the geographical location information of each resource in the virtual power plant to obtain a data set G, and generating a neighborhood radius list according to the data set G;

[0057] In this step, the specific implementation steps of generating a neighborhood radius list according to the data set G include step S211 and step S212;

[0058] Step S211, calculating the distance distribution matrix of each geographical location information in the data set;

[0059] The specific implementation steps of this step include step S2111;

[0060] Step S2111: Calculate the distance distribution matrix of each geographical location information in the data set according to formula (1), wherein formula (1) is:

[0061] G n×n ={Point_Dist(i,j)1≤i≤n,1≤j≤n} (1)

[0062] In formula (1), G n×n is an n×n symmetric matrix; n is the length of the data set G; Point_Dist(i,j) is the distance between the i-th geographic location information and the j-th geographic location information in the data set G.

[0063] Step S212, sort the data in each row of the distance distribution matrix from small to large, and after sorting, use the quartile method to remove isolated points from each column of the distance distribution matrix, and then calculate the expectation of the retained data points to obtain the first value; save all the calculated first values ​​in the neighborhood radius list.

[0064] Step S22, generating a neighborhood minimum point list according to the neighborhood radius list; adjusting parameters of a density-based clustering algorithm according to the neighborhood radius list and the neighborhood minimum point list to obtain an adjusted clustering algorithm.

[0065] The specific implementation steps of this step include step S221 and step S222;

[0066] Step S221: for all sample points contained in the neighborhood corresponding to each neighborhood radius in the neighborhood radius list, use their mathematical expectation value as a parameter in the neighborhood minimum point list to obtain the neighborhood minimum point list;

[0067] This step can be understood as:

[0068]

[0069] In formula (2), Pi is the number of samples in the Epsilon neighborhood of the i-th sample point, and n is the total number of sample points in the data set G.

[0070] Step S222: arbitrarily select a value from the neighborhood radius list and the neighborhood minimum point number list as a parameter in the density-based clustering algorithm, and after selection, perform clustering analysis using the density-based clustering algorithm to obtain multiple third clustering results, and obtain an adjusted clustering algorithm based on the multiple third clustering results.

[0071] In this step, the specific implementation steps of obtaining the adjusted clustering algorithm based on the multiple third clustering results include step S2221;

[0072] Step S2221, using the silhouette coefficient method to calculate the silhouette coefficients corresponding to multiple third clustering results; analyzing the silhouette coefficient, wherein if the silhouette coefficient is greater than a preset threshold, the currently selected neighborhood radius and the minimum number of neighborhood points are used as final parameters of the density-based clustering algorithm, otherwise, re-selecting a value from the neighborhood radius list and the minimum number of neighborhood points list as a parameter in the density-based clustering algorithm, and performing clustering and silhouette coefficient calculation again until the silhouette coefficient is greater than the preset threshold.

[0073] The traditional DBSCAN clustering algorithm has a lot of human factors in the training process that affect the clustering effect. Through the above steps, a DBSCAN algorithm is proposed that can adaptively find the optimal neighborhood radius Epsilon parameter and the optimal neighborhood minimum point number MinPts parameter without human intervention. This algorithm makes up for the shortcomings of the traditional DBSCAN clustering algorithm.

[0074] After the above clustering, considering that the resources contained in each partition are preferably evenly distributed to avoid uneven distribution, this embodiment adjusts each clustering result after clustering, and the adjustment steps can make the resources of each partition more evenly distributed;

[0075] Step S3, count the number of geographic location information contained in each first clustering result, adjust the first clustering result according to the number, and obtain multiple second clustering results, the number of geographic location information contained in each second clustering result is the same, and take the range formed by each second clustering result as a partition.

[0076] The specific implementation steps of this step include step S31;

[0077] Step S31, counting the number of geographic location information included in each first clustering result, if the number is not equal to a preset value, adjusting the geographic location information included in each first clustering result, so that the number of geographic location information included in each first clustering result is equal to the preset value after adjustment.

[0078] The preset value in this step can be customized according to the needs of the user, and the specific implementation steps include step S311;

[0079] Step S311: record the first clustering result whose number of geographic location information is equal to a preset value as a cluster that does not need to be adjusted; record the first clustering result whose number of geographic location information is not equal to the preset value as a cluster that needs to be adjusted; calculate the distance from each cluster that needs to be adjusted to each cluster that does not need to be adjusted, add the distances corresponding to each cluster that needs to be adjusted to obtain a first value, start traversing from the cluster that needs to be adjusted with the largest first value, and make adjustments so that the number of geographic location information contained in each first clustering result is equal to the preset value.

[0080] The specific implementation steps of this step include step S3111;

[0081] Step S3111, start traversing from the cluster that needs to be adjusted with the largest first value. When the number of geographic location information contained in the cluster that needs to be adjusted with the largest first value is less than the preset value, add the geographic location information closest to the centroid of the cluster that needs to be adjusted with the largest first value to the cluster that needs to be adjusted with the largest first value, so that the number of geographic location information contained in the cluster that needs to be adjusted with the largest first value is equal to the preset value; when the number of geographic location information contained in the cluster that needs to be adjusted with the largest first value is greater than the preset value, start with the geographic location information closest to the centroid of all geographic location information in the cluster that needs to be adjusted with the largest first value, and remove them from the cluster that needs to be adjusted with the largest first value in turn, until the number of geographic location information contained in the cluster that needs to be adjusted with the largest first value is equal to the preset value; at the same time, during the adjustment process, when the number of geographic location information contained is equal to the preset value, the first clustering result is recorded as a cluster that does not need to be adjusted.

[0082] In step S3, the range formed by each second clustering result is taken as a partition, and the specific implementation steps include step S32;

[0083] Step S32: The geographical range formed on the map by all the geographical location information contained in each second clustering result is taken as a partition.

[0084] Step S4: Input the geographical location information of each resource into the pre-trained multi-layer perceptron to obtain the importance score of each resource, and send the total importance score corresponding to each partition to the staff to help the staff manage each partition.

[0085] In this step, the pre-trained multi-layer perceptron represents the one-to-one correspondence between geographic location information and importance scores. The total importance score can help staff understand the importance of each partition, which is more conducive to management decision-making.

[0086] In this embodiment, the geographical location of each energy resource is first obtained, and a cluster analysis is performed from the perspective of geographical location to form a plurality of clustering results. At the same time, during clustering, the traditional density-based clustering algorithm is improved, and the clustering effect of the improved clustering algorithm is more superior. After clustering, considering the uniformity of resources in each partition, each clustering result is dynamically adjusted, and the resources in each partition can be balanced through the adjustment method. Finally, the importance of each partition resource is scored. This method can help staff understand the importance of each partition, which is more conducive to management and regulation.

[0087] Example 2

[0088] like Figure 2 As shown, this embodiment provides a virtual power plant resource partitioning device based on a multi-layer perceptron, and the device includes an acquisition module 1, a clustering module 2, an adjustment module 3 and a management module 4.

[0089] Acquisition module 1, used to obtain geographical location information of each resource in the virtual power plant;

[0090] Clustering module 2, used to adjust the density-based clustering algorithm according to the geographical location information of each resource in the virtual power plant, and use the adjusted density-based clustering algorithm to perform cluster analysis on the geographical location information of each resource to obtain multiple first clustering results;

[0091] An adjustment module 3 is used to count the number of geographical location information included in each first clustering result, adjust the first clustering result according to the number, obtain multiple second clustering results, each of which includes the same number of geographical location information, and regard the range formed by each second clustering result as a partition;

[0092] Management module 4 is used to input the geographical location information of each resource into a pre-trained multi-layer perceptron to obtain the importance score of each resource, and send the total importance score corresponding to each partition to the staff to help the staff manage each partition.

[0093] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0094] Example 3

[0095] Corresponding to the above method embodiments, the embodiments of the present disclosure also provide a virtual power plant resource partitioning device based on a multi-layer perceptron. The virtual power plant resource partitioning device based on a multi-layer perceptron described below and the virtual power plant resource partitioning method based on a multi-layer perceptron described above can be referenced to each other.

[0096] Figure 3 FIG. 3 is a block diagram of a virtual power plant resource partitioning device 300 based on a multi-layer perceptron according to an exemplary embodiment. Figure 3 As shown, the virtual power plant resource partitioning device 300 based on multi-layer perceptron may include: a processor 301 and a memory 302. The virtual power plant resource partitioning device 300 based on multi-layer perceptron may also include one or more of a multimedia component 303, an I / O interface 304, and a communication component 305.

[0097] The processor 301 is used to control the overall operation of the multi-layer perceptron-based virtual power plant resource partitioning device 300 to complete all or part of the steps in the multi-layer perceptron-based virtual power plant resource partitioning method. The memory 302 is used to store various types of data to support the operation of the multi-layer perceptron-based virtual power plant resource partitioning device 300, and these data may include, for example, instructions for any application or method operating on the multi-layer perceptron-based virtual power plant resource partitioning device 300, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 303 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 302 or sent via the communication component 305. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 304 provides an interface between the processor 301 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 305 is used for wired or wireless communication between the virtual power plant resource partitioning device 300 based on the multi-layer perceptron and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 305 can include: Wi-Fi module, Bluetooth module, NFC module.

[0098] In an exemplary embodiment, the multi-layer perceptron-based virtual power plant resource partitioning device 300 can be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned multi-layer perceptron-based virtual power plant resource partitioning method.

[0099] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the virtual power plant resource partitioning method based on the multilayer perceptron are implemented. For example, the computer-readable storage medium may be the memory 302 including the program instructions, and the program instructions may be executed by the processor 301 of the virtual power plant resource partitioning device 300 based on the multilayer perceptron to complete the virtual power plant resource partitioning method based on the multilayer perceptron.

[0100] Example 4

[0101] Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a readable storage medium. The readable storage medium described below and the virtual power plant resource partitioning method based on the multi-layer perceptron described above can refer to each other.

[0102] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the virtual power plant resource partitioning method based on a multilayer perceptron in the above method embodiment.

[0103] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A virtual power plant resource partitioning method based on multi-layer perceptron, characterized in that: include: Obtain the geographic location information of each resource in the virtual power plant; Adjusting the density-based clustering algorithm according to the geographic location information of each resource in the virtual power plant, and performing cluster analysis on the geographic location information of each resource using the adjusted density-based clustering algorithm to obtain a plurality of first clustering results; Counting the number of geographical location information included in each first clustering result, adjusting the first clustering result according to the number, obtaining a plurality of second clustering results, each of which includes the same number of geographical location information, and treating the range formed by each second clustering result as a partition; The geographical location information of each resource is input into the pre-trained multi-layer perceptron to obtain the importance score of each resource, and the total importance score corresponding to each partition is sent to the staff to help the staff manage each partition.

2. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 1 is characterized in that: The density-based clustering algorithm is adjusted based on the geographic location information of each resource in the virtual power plant, including: The geographical location information of each resource in the virtual power plant is collected to obtain a data set G, and a neighborhood radius list is generated according to the data set G; A neighborhood minimum point list is generated according to the neighborhood radius list; and parameters of a density-based clustering algorithm are adjusted according to the neighborhood radius list and the neighborhood minimum point list to obtain an adjusted clustering algorithm.

3. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 2 is characterized in that: Generate a neighborhood radius list according to the data set G, including: Calculate the distance distribution matrix of each geographic location information in the data set; The data in each row of the distance distribution matrix are sorted from small to large. After sorting, the quartile method is used to remove isolated points from each column of the distance distribution matrix, and then the expectation of the retained data points is calculated to obtain the first value; all the calculated first values ​​are saved in the neighborhood radius list.

4. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 3 is characterized in that: Calculate the distance distribution matrix of each geographic location information in the data set, including: The distance distribution matrix of each geographic location information in the data set is calculated according to formula (1), and the formula (1) is: G n×n ={Point_Dist(i,j)|1≤i≤n,1≤j≤n} (1) In formula (1), G n×n is an n×n symmetric matrix; n is the length of the data set G; Point_Dist(i,j) is the distance between the i-th geographic location information and the j-th geographic location information in the data set G.

5. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 4 is characterized in that: Generate a neighborhood minimum point list according to the neighborhood radius list; The density-based clustering algorithm is parameter-adjusted according to the neighborhood radius list and the neighborhood minimum point number list to obtain an adjusted clustering algorithm, including: For all the sample points contained in the neighborhood corresponding to each neighborhood radius in the neighborhood radius list, the mathematical expectation value is used as a parameter in the neighborhood minimum point list to obtain the neighborhood minimum point list; A value is arbitrarily selected from the neighborhood radius list and the neighborhood minimum point number list as a parameter in the density-based clustering algorithm. After the selection, a clustering analysis is performed using the density-based clustering algorithm to obtain multiple third clustering results, and an adjusted clustering algorithm is obtained based on the multiple third clustering results.

6. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 5 is characterized in that: The adjusted clustering algorithm is obtained based on the multiple third clustering results, including: The silhouette coefficient method is used to calculate the silhouette coefficients corresponding to the multiple third clustering results; the silhouette coefficient is analyzed, wherein if the silhouette coefficient is greater than a preset threshold, the currently selected neighborhood radius and the minimum number of neighborhood points are used as final parameters of the density-based clustering algorithm, otherwise, a value is randomly selected from the neighborhood radius list and the minimum number of neighborhood points list as a parameter in the density-based clustering algorithm, and clustering and silhouette coefficient calculation are performed again until the silhouette coefficient is greater than the preset threshold.

7. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 1 is characterized in that: The number of geographical location information included in each first clustering result is counted, and the first clustering result is adjusted according to the number to obtain multiple second clustering results, including: The number of geographic location information included in each first clustering result is counted, and if the number is not equal to a preset value, the geographic location information included in each first clustering result is adjusted, so that the number of geographic location information included in each first clustering result is equal to the preset value after adjustment.

8. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 7 is characterized in that: The geographical location information included in each first clustering result is adjusted so that the number of geographical location information included in each first clustering result is equal to a preset value after the adjustment, including: The first clustering result whose number of geographic location information is equal to the preset value is recorded as a cluster that does not need to be adjusted; the first clustering result whose number of geographic location information is not equal to the preset value is recorded as a cluster that needs to be adjusted; the distance from each cluster that needs to be adjusted to each cluster that does not need to be adjusted is calculated, and the corresponding distances of each cluster that needs to be adjusted are added to obtain a first value, and traversal is started from the cluster that needs to be adjusted with the largest first value, and adjustments are made so that the number of geographic location information contained in each first clustering result is equal to the preset value.

9. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 8 is characterized in that: Traversing from the cluster to be adjusted with the largest first value, and adjusting, after the adjustment, the number of geographical location information included in each first clustering result is equal to the preset value, including: The traversal is started from the cluster requiring adjustment with the largest first value. When the number of geographic location information contained in the cluster requiring adjustment with the largest first value is less than the preset value, the geographic location information closest to the centroid of the cluster requiring adjustment with the largest first value is added to the cluster requiring adjustment with the largest first value, so that the number of geographic location information contained in the cluster requiring adjustment with the largest first value is equal to the preset value. When the number of geographic location information contained in the cluster requiring adjustment with the largest first value is greater than the preset value, the geographic location information closest to the centroid of all geographic location information in the cluster requiring adjustment with the largest first value is removed from the cluster requiring adjustment with the largest first value in turn, until the number of geographic location information contained in the cluster requiring adjustment with the largest first value is equal to the preset value. Meanwhile, during the adjustment process, when the number of geographic location information contained is equal to the preset value, the first clustering result is recorded as a cluster that does not require adjustment.

10. The virtual power plant resource partitioning method based on multi-layer perceptron according to claim 1 is characterized in that: The range formed by each second clustering result is taken as a partition, including: The geographical range formed on the map by all the geographical location information included in each second clustering result is regarded as a partition.

Images