New energy output scene reduction method and system

Abstract

The invention discloses a new energy output scene reduction method and system, and the method comprises the steps: carrying out the clustering center selection of original scenes in an obtained original scene set according to a density priority principle, obtaining a first initial clustering center, and carrying out the reduction of a new energy output scene based on the first initial clustering center; gradually selecting other initial clustering centers in the original scene by adopting a maximum and minimum distance principle; the maximum and minimum distance principle avoids the situation that initial clustering center selection may be too adjacent, an object with the Euclidean distance as far as possible is selected as the clustering center, the stability of the clustering center is improved, it is ensured that the minimum distance between the newly selected center and all the selected centers is maximized, and the clustering accuracy is improved. Therefore, all the initial clustering centers are uniformly and dispersedly distributed in the data space as far as possible, and the situation that the initial centers are too adjacent in a traditional method is avoided. According to the strategy, the representativeness and stability of the clustering center are effectively improved, and time correlation decoupling caused by improper selection of the initial center is prevented.

Description

Technical Field

[0001] This invention belongs to the field of new energy regulation technology, specifically relating to a method and system for reducing new energy output scenarios. Background Technology

[0002] In the SCUC problem for power systems containing renewable energy sources (mainly wind and solar power), the high volatility and correlation of renewable energy output are core factors affecting the accuracy and security of dispatch optimization. While large-scale scenario generation can comprehensively characterize the stochastic characteristics of wind and solar power output, the resulting computational burden severely impacts the solution of the power system dispatch model. The large number of generated original scenarios leads to an exponential increase in the computational complexity of the power system dispatch model, resulting in excessively long solution times or even unsolvable problems. Therefore, efficient scenario reduction techniques are needed to reduce the computational dimensionality while ensuring that the reduced scenarios maximally represent the statistical characteristics of the original scenarios. Although the traditional K-means clustering algorithm is fast, its random selection of initial cluster centers can lead to algorithm instability and may decouple the temporal correlation of wind and solar power output. Therefore, an improved K-means clustering algorithm is proposed for scenario reduction. By optimizing the initial cluster center selection strategy, the stability and effectiveness of scenario reduction are improved. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for reducing new energy power output scenarios, so as to overcome the problems that existing technologies are prone to algorithm instability due to the random selection of initial cluster centers, and may decouple the time correlation of wind and solar power output. This invention improves the stability and effectiveness of scenario reduction by adopting a density-first strategy and the maximum-minimum distance principle to optimize the cluster center selection strategy.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for reducing the power output of new energy sources, characterized by comprising the following steps: S1, select cluster centers for the original scenes in the acquired original scene set according to the density priority principle to obtain the first initial cluster center; S2, based on the first initial cluster center, uses the maximum-minimum distance principle to gradually select the remaining initial cluster centers in the original scene; S3. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest cluster center belongs. The mean of all scenes in each cluster is taken as the new cluster center. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest new cluster center belongs. The above new cluster center acquisition steps are repeated until the maximum Euclidean distance between two adjacent cluster centers in each cluster is less than a preset threshold. Then the center point of each cluster is finally acquired as a typical scene.

[0005] Preferably, the step of selecting cluster centers for the original scenes in the acquired original scene set according to the density priority principle to obtain the first initial cluster center includes: defining a density parameter by calculating the Euclidean distance between each scene and its neighboring scenes, and selecting the scene with the highest density as the first cluster center.

[0006] Preferably, the Euclidean distance between all scenes in the original scene set is calculated, and the Euclidean distance information is stored in a distance distribution matrix; wherein the Euclidean distance and the distance distribution matrix are represented as follows: (1) (2) in, and Represents the original scene set The scene in; This indicates the number of elements contained in each scene; It is a symmetric matrix with zero diagonal elements; For distance distribution matrix Find the distance parameter for each row Represented as the first A scenario The density is stored in the distance array. In the distance array, the smallest distance parameter This represents the scene with the highest density. That is, the first initial cluster center .

[0007] Preferably, the distance parameters and distance array are represented as follows: (3) (4) in, Operator representation for scene With sets Sort the Euclidean distances of all scenes in the dataset and select the first scene. Output the small Euclidean distance; distance array Stored collection Distance parameters for all scenes in the game.

[0008] Preferably, the stepwise selection of the remaining initial cluster centers in the original scene using the maximum-minimum distance principle specifically includes: The first initial cluster centers were obtained based on the density concept. and the original scene set Total number of clusters ; From the original scene set Find the cluster center with the first initial cluster center. The scene with the largest Euclidean distance was used as the second initial cluster center. ; Calculate the scenario where the cluster center is not selected as the initial cluster center. With the first initial cluster center The second initial cluster center Calculate the Euclidean distance between them and find the scenes that were not selected as initial cluster centers. With the first initial cluster center The second initial cluster center The maximum value of the two Euclidean distances Then the maximum value The corresponding scenario where it was not selected as the initial cluster center This is the third initial cluster center. ; If already available -1 initial cluster centers, calculating scenarios where cluster centers were not selected as initial cluster centers. Calculate the Euclidean distance to each initial cluster center and find the scenarios that were not selected as initial cluster centers. The maximum value of the Euclidean distance to each cluster center Then the maximum value Corresponding scenarios That is, the first Initial cluster centers .

[0009] Preferably, the maximum of the two Euclidean distances. The calculation formula is as follows: (5).

[0010] Preferably, the mean of all scenes within each cluster is used as the new cluster center. As shown in the following formula: (6) in, and They represent the first The number and set of each cluster of scenarios; Repeat the above steps for obtaining new cluster centers until the maximum Euclidean distance between two adjacent cluster centers obtained in each cluster is less than a preset threshold, as shown in the following formula: (7) The maximum Euclidean distance between the cluster centers in two iterations is less than the threshold. Iteration complete.

[0011] In a second aspect, a system for reducing the output of new energy sources is provided, comprising a preprocessing module, a clustering module, and an adjustment module: The preprocessing module selects cluster centers for the original scenes in the acquired set of original scenes according to the density priority principle, and obtains the first initial cluster center. The clustering module, based on the first initial cluster center, uses the maximum-minimum distance principle to progressively select the remaining initial cluster centers in the original scene; The adjustment module assigns all scenes to the cluster to which the nearest cluster center belongs, based on the principle of minimum Euclidean distance. The mean of all scenes in each cluster is used as the new cluster center. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest new cluster center belongs. The above new cluster center acquisition steps are repeated until the maximum Euclidean distance between two adjacent cluster centers in each cluster is less than a preset threshold. Then, the center point of each cluster is finally acquired as a typical scene.

[0012] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for reducing the output of a new energy source scenario.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method for reducing the output of a new energy source.

[0014] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a method for reducing new energy output scenarios. It involves selecting cluster centers from an acquired set of original scenarios according to a density-first principle to obtain an initial cluster center. Based on this initial cluster center, the remaining initial cluster centers in the original scenarios are selected progressively using a maximum-minimum distance principle. Then, based on the minimum Euclidean distance principle, all scenarios are assigned to the cluster to which their nearest cluster center belongs. The average value of all scenarios in each cluster is used as a new cluster center. Again, based on the minimum Euclidean distance principle, all scenarios are assigned to the cluster to which their nearest new cluster center belongs. The remaining initial cluster centers are then selected progressively using the maximum-minimum distance principle. The maximum-minimum distance principle avoids the possibility of initially selecting cluster centers that are too close together, selecting objects with the greatest possible Euclidean distance as cluster centers, thus improving the stability of the cluster centers. It ensures that the minimum distance between the newly selected center and all previously selected centers is maximized, thereby making the distribution of all initial cluster centers as uniform and dispersed as possible in the data space, avoiding the possibility of initially being too close together in traditional methods. This strategy effectively improves the representativeness and stability of the cluster centers and prevents temporal decoupling caused by improper initial center selection. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a method for reducing the output of new energy sources in an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of a new energy output reduction system structure in an embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] like Figure 1 As shown, this invention provides a method for reducing new energy output scenarios. By employing a density-first strategy and the maximum-minimum distance principle to optimize the cluster center selection strategy, the stability and effectiveness of scenario reduction are improved. Specifically, the method includes the following steps: S1, select cluster centers for the original scenes in the acquired original scene set according to the density priority principle to obtain the first initial cluster center; S2, based on the first initial cluster center, uses the maximum-minimum distance principle to gradually select the remaining initial cluster centers in the original scene; S3. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest cluster center belongs. The mean of all scenes in each cluster is taken as the new cluster center. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest new cluster center belongs. The above new cluster center acquisition steps are repeated until the maximum Euclidean distance between two adjacent cluster centers in each cluster is less than a preset threshold. Then the center point of each cluster is finally acquired as a typical scene.

[0020] In a specific embodiment of this application, cluster centers are selected from the obtained original scene set according to a density-first principle to obtain the first initial cluster center. This is achieved by using cluster analysis to reduce the generated original scene set according to certain criteria, which can improve the stochastic optimization of the power system. The scene with the highest density among all the original scenes is selected as the first initial cluster center. The density parameter is defined by calculating the Euclidean distance between each scene and its neighboring scenes, and the scene with the highest density is selected as the first cluster center. This ensures that the first cluster center is the most representative high-density region in the original dataset, laying a stable foundation for subsequent clustering and avoiding the instability that may be caused by random selection.

[0021] Cluster centers are selected according to the density-first principle, specifically including: Based on the original scene set Calculate the Euclidean distance between each scene and its m nearest neighbor scenes. Set the density parameter to the maximum value among these minimum distances, representing the density of the space in which the point is located. The smaller the density parameter, the denser the region where the scene point is located.

[0022] Calculate the original scene set The Euclidean distances between all scenes within the area are calculated, and the Euclidean distance information is stored in a distance distribution matrix; the Euclidean distances and the distance distribution matrix are represented as follows: (1) (2) in, and Represents the original scene set The scene in; This indicates the number of elements contained in each scene. Assuming that one data point is taken every 15 minutes, then 96 data points can be taken in one cycle (24 hours). It is a symmetric matrix with zero diagonal elements.

[0023] For distance distribution matrix Find the distance parameter for each row Represented as the first A scenario The density is stored in the distance array. In the distance array, the smallest distance parameter This represents the scene with the highest density. That is, the first initial cluster center The distance parameters and distance array are represented as follows: (3) (4) in, Operator representation for scene With sets Sort the Euclidean distances of all scenes in the dataset and select the first scene. Output the small Euclidean distance; distance array Stored collection Distance parameters for all scenes in the game.

[0024] In a specific embodiment of the present invention, after determining the first cluster center, the remaining initial cluster centers are selected step by step using the maximum-minimum distance principle. The maximum-minimum distance principle avoids the possibility of initially selecting cluster centers that are too close together, selecting objects with the greatest possible Euclidean distance as cluster centers, thus improving the stability of the cluster centers. It ensures that the minimum distance between the newly selected center and all previously selected centers is maximized, thereby making all initial cluster centers as evenly and dispersedly distributed in the data space as possible, avoiding the possibility of initially being too close together in traditional methods. This strategy effectively improves the representativeness and stability of the cluster centers and prevents temporal decoupling caused by improper selection of initial centers.

[0025] The specific process of obtaining the remaining initial cluster centers using the maximum-minimum distance principle includes: The first initial cluster centers were obtained based on the density concept. and the original scene set Total number of clusters ; From the original scene set Find the cluster center with the first initial cluster center. The scene with the largest Euclidean distance was used as the second initial cluster center. ; Calculate the scenario where the cluster center is not selected as the initial cluster center. With the first initial cluster center The second initial cluster center Calculate the Euclidean distance between them and find the scenes that were not selected as initial cluster centers. With the first initial cluster center The second initial cluster center The maximum value of the two Euclidean distances Then the maximum value The corresponding scenario where it was not selected as the initial cluster center This is the third initial cluster center. . The calculation formula is as follows: (5) If already available -1 initial cluster centers, calculating scenarios where cluster centers were not selected as initial cluster centers. Calculate the Euclidean distance to each initial cluster center and find the scenarios that were not selected as initial cluster centers. The maximum value of the Euclidean distance to each cluster center Then the maximum value Corresponding scenarios That is, the first Initial cluster centers .

[0026] After determining the initial cluster centers, all original scenes are assigned to various clusters according to the principle of minimum Euclidean distance.

[0027] The original scene set is based on the principle of minimum Euclidean distance. After all scenes are assigned to the cluster to which the nearest cluster center belongs, the mean of all scenes within each cluster is used as the new cluster center. , collect the original scene All scenes are assigned to the new cluster center that is closest to them. Repeat this step within the relevant cluster until the maximum Euclidean distance between two adjacent cluster centers in each cluster is less than a preset threshold. Then, the center point of each cluster obtained at the end becomes a typical scenario.

[0028] The mean of all scenes within each cluster is used as the new cluster center. As shown in the following formula: (6) in, and They represent the first The number and set of each cluster scenario.

[0029] The maximum Euclidean distance between two consecutive cluster centers obtained in each cluster is less than a preset threshold, as shown in the following formula: (7) The maximum Euclidean distance between the cluster centers in two iterations is less than the threshold. The iteration has ended, and the threshold value in this application is... .

[0030] The center point of each cluster becomes a typical scenario. To ensure probability consistency, the typical scenario of each cluster is assigned the following probabilities: (8) in, Indicates the first The probability of a typical scenario For the first The number of original scenes contained in each cluster This represents the total number of scenes.

[0031] Example 1: Input the original scenario set S: Given a dataset containing n new energy power output scenarios. Each scene It contains t data points (e.g., one data point every 15 minutes within 24 hours, for a total of 96 data points).

[0032] For each scenario in set S Calculate the Euclidean distance between the scene and its m nearest neighbors; define the maximum of these minimum distances as the density parameter of the scene. The smaller the density parameter, the denser the region in which the scene is located.

[0033] Specifically, first calculate any two scenarios and Euclidean distance between them: .

[0034] Then, the Euclidean distances between all scenes are stored in the distance distribution matrix: .

[0035] Next, for each scenario The Euclidean distances from the m-th scene to all other scenes are sorted, and the m-th smallest Euclidean distance is selected as its density parameter. The density parameters of all scenes are stored in the distance array.

[0036] Determine the first initial cluster center From the distance array Find the minimum density parameter in the middle. The scenario corresponding to this parameter That is, selected as the first initial cluster center. ; The maximum-minimum distance principle is used to progressively select the remaining initial cluster centers: Select the second initial cluster center In determining Then, find the matching set from set S. The scenario with the largest Euclidean distance is used as the second initial cluster center. After selecting clusters using the method described above, continue to obtain the initial cluster centers. to For the third initial cluster center : Calculate all scenarios that were not selected as initial cluster centers With the first initial cluster center and the second initial cluster center The Euclidean distance between them. Then, find With the first initial cluster center and the second initial cluster center Maximum value of those that are relatively close in distance ;choose The largest scene as Similarly, if there are already k-1 initial cluster centers, then calculate all scenarios that were not selected as initial cluster centers. Find the Euclidean distance between each initial cluster center and the Euclidean distance. Maximum value of minimum distance from each cluster center The corresponding scenario That is, the kth initial cluster center .

[0037] Iterative clustering and probability weight allocation: Determine k initial cluster centers ,..., Then, all scenes in set S are assigned to the clusters belonging to the nearest cluster centers according to the principle of minimum Euclidean distance; the mean of all scenes in each cluster is calculated and used as the new cluster centers. Repeat this step until the maximum Euclidean distance between the new cluster center and the previous cluster center in each cluster is less than a preset threshold ε, at which point the iteration ends; after clustering, the center point of each cluster becomes a typical scenario. To ensure probability consistency, for each typical scenario... Assign probability weights, which are equal to the number of original scenes included in the cluster. Divide by the total number of scenes, n.

[0038] This invention, through the combination of the two strategies mentioned above, achieves more stable and representative initial cluster centers before iterative calculations, thereby improving the stability and effectiveness of the entire scenario reduction process. Ultimately, it can more accurately characterize the statistical characteristics of new energy output and effectively reduce the computational complexity and solution difficulty of the power system dispatch model. This invention improves the stability and effectiveness of scenario reduction by employing a density-first strategy and the maximum-minimum distance principle to optimize the cluster center selection strategy. First, the density-first strategy is used to select the first cluster center. After determining the first cluster center, the maximum-minimum distance principle is used to progressively select the remaining initial cluster centers. The maximum-minimum distance principle avoids the possibility of initially selecting cluster centers that are too close together, selecting objects with the greatest possible Euclidean distance as cluster centers, thus improving the stability of the cluster centers.

[0039] In another embodiment of the present invention, such as Figure 2 As shown, a system for reducing power output in new energy scenarios is provided, characterized by including a preprocessing module, a clustering module, and an adjustment module: The preprocessing module selects cluster centers for the original scenes in the acquired set of original scenes according to the density priority principle, and obtains the first initial cluster center. The clustering module, based on the first initial cluster center, uses the maximum-minimum distance principle to progressively select the remaining initial cluster centers in the original scene; The adjustment module assigns all scenes to the cluster to which the nearest cluster center belongs, based on the principle of minimum Euclidean distance. The mean of all scenes in each cluster is used as the new cluster center. Based on the principle of minimum Euclidean distance, all scenes are assigned to the cluster to which the nearest new cluster center belongs. The above new cluster center acquisition steps are repeated until the maximum Euclidean distance between two adjacent cluster centers in each cluster is less than a preset threshold. Then, the center point of each cluster is finally acquired as a typical scene.

[0040] In another embodiment of the present invention, a terminal device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a method for reducing the output of new energy sources.

[0041] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). This computer-readable storage medium is a memory device in a terminal device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device. The computer-readable storage medium provides storage space containing the terminal's operating system. Furthermore, this storage space also contains one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the new energy output reduction method in the above embodiments.

[0042] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0043] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0044] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0045] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for reducing new energy output scenarios, characterized in that, The method comprises the following steps: The first initial clustering center is obtained by selecting the clustering center of the original scene in the obtained original scene set according to the density-first principle. The remaining initial clustering centers in the original scene are selected step by step based on the maximum-minimum distance principle based on the first initial clustering center. All scenes are assigned to the class cluster to which the nearest clustering center belongs based on the minimum Euclidean distance principle, the mean value of all scenes in each class cluster is taken as a new clustering center, and all scenes are assigned to the class cluster to which the nearest new clustering center belongs based on the minimum Euclidean distance principle. The new clustering center acquisition step is repeated until the maximum Euclidean distance between the clustering centers obtained in adjacent two times in each class cluster is less than a preset threshold, and the center point of each class cluster finally obtained becomes a typical scene.

2. The new energy output scene reduction method according to claim 1, characterized in that, The first initial clustering center is obtained by selecting the clustering center of the original scene in the obtained original scene set according to the density-first principle.

3. The new energy output scene reduction method according to claim 2, characterized in that, The Euclidean distance between all scenes in the original scene set is calculated, and the Euclidean distance information is stored in a distance distribution matrix; wherein the Euclidean distance and the distance distribution matrix are represented as follows: （1） （2） wherein, and denotes the original set of scenes ; denotes the number of elements contained in each scene; is a symmetric matrix with zero diagonal elements; to distance distribution matrix Each row finds distance parameter Represented as the Scene Density and stored in distance array The smallest distance parameter in distance array Represents the scene with the most density That is the first initial cluster center .

4. The new energy output scene reduction method according to claim 3, characterized in that, The distance parameter and the distance array are represented as follows: （3） （4） in, Operator representation for scene With sets Sort the Euclidean distances of all scenes in the dataset and select the first scene. Output the small Euclidean distance; distance array Stored collection Distance parameters for all scenes in the game.

5. The new energy output scene reduction method according to claim 1, characterized in that, The remaining initial clustering centers in the original scene are selected step by step based on the maximum-minimum distance principle, specifically including: The first initial clustering center obtained based on the density idea and the original scene set The total number of category clusters ; From the original scene set Find the cluster center with the first initial cluster center. The scene with the largest Euclidean distance was used as the second initial cluster center. ; Calculate the scenario where the cluster center is not selected as the initial cluster center. With the first initial cluster center The second initial cluster center Calculate the Euclidean distance between them and find the scenes that were not selected as initial cluster centers. With the first initial cluster center The second initial cluster center The maximum value of the two Euclidean distances Then the maximum value The corresponding scenario where it was not selected as the initial cluster center This is the third initial cluster center. ; If already available -1 initial cluster centers, calculating scenarios where cluster centers were not selected as initial cluster centers. Calculate the Euclidean distance to each initial cluster center and find the scenarios that were not selected as initial cluster centers. The maximum value of the Euclidean distance to each cluster center Then the maximum value Corresponding scenarios That is, the first Initial cluster centers .

6. The new energy output scene reduction method according to claim 5, characterized in that, the maximum of the two euclidean distances The formula for calculating the maximum of the two euclidean distances is as follows: （5）。 7. The new energy output scene reduction method according to claim 1, characterized in that, said taking the mean of all scenes within each cluster as a new cluster center as shown in the following equation: （6） wherein, and denote the number of clusters and the set of the first cluster scenarios, respectively. The new clustering center acquisition step is repeated until the maximum Euclidean distance between the clustering centers obtained in adjacent two times in each class cluster is less than a preset threshold, as shown in the following formula: （7） The maximum of the Euclidean distance of the two iterations of each cluster center is less than a threshold value The iterations are finished.

8. A new energy output scene reduction system, characterized in that, The method comprises a preprocessing module, a clustering module and an adjustment module: The first initial clustering center is obtained by selecting the clustering center of the original scene in the obtained original scene set according to the density-first principle. The remaining initial clustering centers in the original scene are selected step by step based on the maximum-minimum distance principle based on the first initial clustering center. All scenes are assigned to the class cluster to which the nearest clustering center belongs based on the minimum Euclidean distance principle, the mean value of all scenes in each class cluster is taken as a new clustering center, and all scenes are assigned to the class cluster to which the nearest new clustering center belongs based on the minimum Euclidean distance principle. The new clustering center acquisition step is repeated until the maximum Euclidean distance between the clustering centers obtained in adjacent two times in each class cluster is less than a preset threshold, and the center point of each class cluster finally obtained becomes a typical scene.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the new energy output scene reduction method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to realize the steps of the new energy output scene reduction method according to any one of claims 1 to 7.

Images