New energy power evaluation method and system based on K-means clustering

Abstract

The invention provides a new energy power evaluation method and system based on K-means clustering. The method comprises the steps of obtaining historical power generation data of a new energy power station; setting a clustering number K according to a new energy power evaluation requirement, and clustering the historical power generation data through a K-means clustering algorithm to obtain k clustering centers and k + 1 power generation capability levels; based on the k clustering centers and the k + 1 power generation capacity grades, new energy power evaluation background plates of different areas and different seasons are drawn; and putting the new energy power generation prediction curve to be evaluated into the corresponding new energy power evaluation background board to realize evaluation of the new energy power generation capacity. A scheduling mechanism can be helped to take measures in advance to stabilize a large new energy fluctuation process, and safe and reliable operation of a power system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of renewable energy power generation, and in particular to a renewable energy power evaluation method and system based on K-means clustering. Background Art

[0002] As the proportion of renewable energy in the power system increases, the randomness and volatility of its output have brought greater challenges to the balance of the power system and the absorption of renewable energy. When wind and light resources change sharply or there is continuous low wind and low light weather, both the supply and demand sides of the power system face strong uncertainty, and the risk of power abandonment and power shortage coexist. Due to the different power load demands in different regions, seasons and time periods, the same wind and light resources have different effects on the balance and absorption of the system. For example, the power load demand is small at night, and the wind power generation capacity under certain wind conditions may be relatively sufficient relative to the power demand, but the power demand during the evening peak period is much higher than that at night, and the wind power generation capacity under the same wind conditions is usually difficult to meet the power demand. Although there are currently a variety of renewable energy power generation prediction models for estimating the renewable energy power generation capacity in the future, it is particularly important to reasonably evaluate the impact of renewable energy power generation capacity on system balance and absorption in the future, and help dispatching agencies take measures in advance to smooth out the large renewable energy fluctuation process. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a new energy power evaluation method based on K-means clustering, comprising:

[0004] Obtain historical power generation data of new energy power plants;

[0005] The number of clusters K is set according to the new energy power evaluation requirements, and the historical power generation data is clustered by a K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels;

[0006] Based on the k cluster centers and k+1 power generation capacity levels, new energy power evaluation background boards for different regions and seasons are drawn; the new energy power generation prediction curve to be evaluated is placed in the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity.

[0007] Furthermore, after the step of obtaining the historical power generation data of the new energy power station, it also includes:

[0008] The historical power generation data is cleaned, and the cleaned data is normalized according to evaluation requirements.

[0009] Furthermore, the historical power generation data is cleaned and the cleaned data is normalized according to the evaluation requirements, including:

[0010] Find missing values ​​in the data and fill them in through linear interpolation;

[0011] Filter outliers in the data and correct them by calculating the average between two points;

[0012] The cleaned data is normalized according to the evaluation requirements of different regions, seasons and time periods.

[0013] Furthermore, the number of clusters K is set according to the new energy power evaluation requirements, and the historical power generation data is clustered by the K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels, including:

[0014] The number of clusters K is set according to the new energy power evaluation requirements;

[0015] One-dimensional search method is used to randomly generate multiple groups of initial cluster centers;

[0016] The normalized historical power generation data of new energy power stations are clustered using the initial cluster centers to obtain k cluster centers and k+1 power generation capacity levels.

[0017] Furthermore, based on the k cluster centers and k+1 power generation capacity levels, a new energy power evaluation background board for different regions and seasons is drawn, including:

[0018] Based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, evaluation background boards for different regions and seasons are drawn with time as the horizontal axis and 0-1 as the vertical axis.

[0019] Furthermore, the prediction curve of the new energy power generation to be evaluated is placed in the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity, including:

[0020] Normalize the prediction curve of renewable energy power generation to be evaluated and put it into the corresponding evaluation background board to realize the evaluation of renewable energy power generation capacity.

[0021] The present invention also provides a new energy power evaluation system based on K-means clustering, comprising:

[0022] The power generation data acquisition module is used to obtain the historical power generation data of the new energy power station;

[0023] A clustering module is used to set the number of clusters K according to the new energy power evaluation requirements, and cluster the historical power generation data through a K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels;

[0024] The evaluation module is used to draw new energy power evaluation background boards in different regions and seasons based on the k cluster centers and k+1 power generation capacity levels; the new energy power generation prediction curve to be evaluated is placed in the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity.

[0025] Furthermore, it also includes:

[0026] The normalization processing module is used to clean the historical power generation data and perform normalization processing on the cleaned data according to evaluation requirements.

[0027] Furthermore, the clustering module includes:

[0028] The cluster number setting submodule is used to set the cluster number K according to the new energy power evaluation requirements;

[0029] The initial cluster center generation submodule is used to randomly generate multiple groups of initial cluster centers using a one-dimensional search method;

[0030] The clustering submodule is used to cluster the normalized historical power generation data of new energy power stations using the initial clustering centers to obtain k clustering centers and k+1 power generation capacity levels.

[0031] Furthermore, the evaluation module includes:

[0032] The evaluation background board drawing submodule is used to draw the evaluation background boards of different regions and seasons based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, with time as the horizontal axis and 0-1 as the vertical axis.

[0033] The present invention provides a new energy power evaluation method system based on K-means clustering. Aiming at the problem of how to reasonably evaluate the impact of new energy power generation capacity on system balance and absorption, the system collects historical power generation data of new energy power stations and determines the classification results of new energy power generation capacity in different regions and seasons based on the K-means clustering algorithm. This realizes an efficient evaluation of the impact of new energy power generation capacity on the system in the future, which is conducive to helping dispatching agencies take measures in advance to smooth out larger new energy fluctuations and ensure safe and reliable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a new energy power evaluation method based on K-means clustering provided by an embodiment of the present invention;

[0035] Figure 2 is an evaluation result of the renewable energy power generation capacity of a certain region on a certain day in spring involved in an embodiment of the present invention;

[0036] Figure 3is an evaluation result of the renewable energy power generation capacity of a certain region on a certain day in winter involved in an embodiment of the present invention;

[0037] Figure 4 It is a structural schematic diagram of a new energy power evaluation system based on K-means clustering provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention, so the present invention is not limited to the specific implementation disclosed below.

[0039] Example 1

[0040] Figure 1 The present invention provides a flow chart of a new energy power evaluation method based on K-means clustering. The method is based on the historical power generation data of regional new energy power stations, pre-sets the number of clusters K according to actual needs, calculates the cluster center through the K-means clustering algorithm, and evaluates the impact of new energy power generation capacity on system balance and consumption in the future based on the clustering results, so as to achieve an efficient evaluation of the impact of new energy power generation capacity on the system. Specifically, it includes the following steps:

[0041] Step S101, obtaining historical power generation data of the new energy power station.

[0042] The historical power generation data of the new energy power station in the past three years is collected in steps of 5 minutes or 15 minutes, and the historical power generation data is cleaned. The cleaned data is normalized according to the evaluation requirements. Specifically, it includes finding missing values ​​in the data and filling the missing values ​​through linear interpolation; screening outliers in the data and correcting the outliers by calculating the average value between two points; and normalizing the cleaned data according to the evaluation requirements of different regions, seasons, and time periods.

[0043] Step S102, setting the number of clusters K according to the new energy power evaluation requirements, clustering the historical power generation data through the K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels.

[0044] The number of clusters K is set according to the evaluation requirements of renewable energy power. In order to eliminate the influence of the initial cluster center on the grading results, a one-dimensional search method is used to randomly generate multiple groups of initial cluster centers. The initial cluster centers are used to cluster the normalized historical power generation data of renewable energy power stations to obtain new cluster centers. The above steps are repeated until the cluster centers no longer change significantly or the preset number of iterations is reached, at which time the final k cluster centers and k+1 power generation capacity levels are obtained.

[0045] Step S103, based on the k cluster centers and k+1 power generation capacity levels, draw new energy power evaluation background boards for different regions and seasons; put the new energy power generation prediction curve to be evaluated into the corresponding new energy power evaluation background board to realize the evaluation of new energy power generation capacity.

[0046] Based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, evaluation background boards for different regions and seasons are drawn with time as the horizontal axis and 0-1 as the vertical axis.

[0047] Normalize the prediction curve of renewable energy power generation to be evaluated and put it into the corresponding evaluation background board to realize the evaluation of renewable energy power generation capacity.

[0048] Example 2

[0049] First, at 15-minute intervals, the historical power generation data series of new energy power stations in different seasons of a regional power grid in the past three years are collected, and the data are preprocessed to obtain the normalized data series. Then, according to the demand, the number of clusters for different time periods is preset to 3. The historical power generation data series of new energy power stations in different seasons and time periods are clustered through the K-means clustering algorithm, and the three cluster centers of different seasons and time periods are calculated (taking spring and winter as examples, the calculation method for other seasons is the same, and the clustering results are shown in Table 1). The three cluster centers correspond to four power generation capacity levels, namely: Class A corresponds to 1~k1, Class B corresponds to k1~k2, Class C corresponds to k2~k3, and Class D corresponds to k3~0. Finally, based on the clustering results of different time periods, an evaluation background board is drawn. Normalize the prediction curve of new energy power generation to be evaluated and put it into the corresponding evaluation background board (see Figure 2 and Figure 3 ), by comparing the position of the predicted power curve of renewable energy power generation in the background board, the renewable energy power evaluation can be realized.

[0050] Table 1 Cluster centers of a regional power grid at different times in spring and winter (%)

[0051]

[0052] Depend on Figure 2 It can be seen that this day is a high-power day in spring as a whole. The new energy power generation capacity during the early and late peak load periods was evaluated as B and A respectively, which is a medium-to-high level of new energy power generation capacity during the early and late peak load periods in the region in spring in the past three years. The new energy power generation capacity in other periods was evaluated as B, which is a medium level in other periods in the spring of the region in the past three years. The amount of new energy power throughout the day was large, and there was pressure to absorb new energy. The dispatching agency should take timely measures such as reducing the output of conventional power sources such as thermal power and actively organizing electricity trading to promote the absorption of new energy.

[0053] Depend on Figure 3 It can be seen that this day is a low-output day in winter. The new energy power generation capacity during the early and late peak load periods was evaluated as Class C, which is a relatively low level of new energy power generation capacity during the early and late peak load periods in the region in winter in the past three years; the new energy power generation capacity in other periods was evaluated as Class C or Class D, which is a low output or relatively low output level in other periods in the region in winter in the past three years. The amount of new energy power throughout the day was relatively small, and there was a risk of insufficient new energy support capacity. The dispatching agency should promptly take measures such as increasing the output of conventional power sources such as thermal power and applying for support from interconnection lines to ensure reliable power supply in the region.

[0054] Depend on Figure 2 and Figure 3 It can be seen that the ratings of the same renewable energy generation capacity in different seasons and different periods of the same season are different, and the degree of influence on system balance and absorption is different. Based on the K-means clustering algorithm, the present invention can achieve efficient evaluation of renewable energy generation power in the future by grading renewable energy generation capacity in different regions and seasons, which is conducive to helping dispatching agencies take measures in advance to smooth out large renewable energy fluctuations and ensure safe and reliable operation of the power system.

[0055] Example 3

[0056] Based on the same inventive concept, the present invention also provides a new energy power evaluation system based on K-means clustering, such as Figure 4 As shown, including:

[0057] The power generation data acquisition module 410 is used to acquire the historical power generation data of the new energy power station;

[0058] The clustering module 420 is used to set the number of clusters K according to the new energy power evaluation requirements, and cluster the historical power generation data through the K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels;

[0059] The evaluation module 430 is used to draw the new energy power evaluation background board in different regions and seasons based on the k cluster centers and k+1 power generation capacity levels; put the new energy power generation prediction curve to be evaluated into the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity.

[0060] Furthermore, it also includes:

[0061] The normalization processing module is used to clean the historical power generation data and perform normalization processing on the cleaned data according to evaluation requirements.

[0062] Furthermore, the clustering module includes:

[0063] The cluster number setting submodule is used to set the cluster number K according to the new energy power evaluation requirements;

[0064] The initial cluster center generation submodule is used to randomly generate multiple groups of initial cluster centers using a one-dimensional search method;

[0065] The clustering submodule is used to cluster the normalized historical power generation data of new energy power stations using the initial clustering centers to obtain k clustering centers and k+1 power generation capacity levels.

[0066] Furthermore, the evaluation module includes:

[0067] The evaluation background board drawing submodule is used to draw the evaluation background boards of different regions and seasons based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, with time as the horizontal axis and 0-1 as the vertical axis.

[0068] The present invention provides a new energy power evaluation method system based on K-means clustering. Aiming at the problem of how to reasonably evaluate the impact of new energy power generation capacity on system balance and absorption, the system collects historical power generation data of new energy power stations and determines the classification results of new energy power generation capacity in different regions and seasons based on the K-means clustering algorithm. This realizes an efficient evaluation of the impact of new energy power generation capacity on the system in the future, which is conducive to helping dispatching agencies take measures in advance to smooth out larger new energy fluctuations and ensure safe and reliable operation of the power system.

[0069] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented 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.

[0070] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0071] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0072] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalents that do not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A new energy power evaluation method based on K-means clustering, characterized in that: include: Obtain historical power generation data of new energy power plants; The number of clusters K is set according to the new energy power evaluation requirements, and the historical power generation data is clustered by a K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels; Based on the k cluster centers and k+1 power generation capacity levels, new energy power evaluation background boards for different regions and seasons are drawn; the new energy power generation prediction curve to be evaluated is placed in the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity.

2. The method according to claim 1, characterized in that After the step of obtaining the historical power generation data of the new energy power station, it also includes: The historical power generation data is cleaned, and the cleaned data is normalized according to evaluation requirements.

3. The method according to claim 2, characterized in that The historical power generation data is cleaned and the cleaned data is normalized according to the evaluation requirements, including: Find missing values ​​in the data and fill them in through linear interpolation; Filter outliers in the data and correct them by calculating the average between two points; The cleaned data is normalized according to the evaluation requirements of different regions, seasons and time periods.

4. The method according to claim 1, characterized in that The number of clusters K is set according to the new energy power evaluation requirements, and the historical power generation data is clustered by the K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels, including: The number of clusters K is set according to the new energy power evaluation requirements; One-dimensional search method is used to randomly generate multiple groups of initial cluster centers; The normalized historical power generation data of new energy power stations are clustered using the initial cluster centers to obtain k cluster centers and k+1 power generation capacity levels.

5. The method according to claim 1, characterized in that: Based on the k cluster centers and k+1 power generation capacity levels, a new energy power evaluation background board for different regions and seasons is drawn, including: Based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, evaluation background boards for different regions and seasons are drawn with time as the horizontal axis and 0-1 as the vertical axis.

6. The method according to claim 1, characterized in that The prediction curve of renewable energy power generation to be evaluated is placed in the corresponding renewable energy power evaluation background board to evaluate renewable energy power generation capacity, including: Normalize the prediction curve of renewable energy power generation to be evaluated and put it into the corresponding evaluation background board to realize the evaluation of renewable energy power generation capacity.

7. A new energy power evaluation system based on K-means clustering, characterized in that: include: The power generation data acquisition module is used to obtain the historical power generation data of the new energy power station; A clustering module is used to set the number of clusters K according to the new energy power evaluation requirements, and cluster the historical power generation data through a K-means clustering algorithm to obtain k cluster centers and k+1 power generation capacity levels; The evaluation module is used to draw new energy power evaluation background boards in different regions and seasons based on the k cluster centers and k+1 power generation capacity levels; the new energy power generation prediction curve to be evaluated is placed in the corresponding new energy power evaluation background board to realize the evaluation of the new energy power generation capacity.

8. The system according to claim 7, characterized in that Also includes: The normalization processing module is used to clean the historical power generation data and perform normalization processing on the cleaned data according to evaluation requirements.

9. The system according to claim 7, characterized in that Clustering modules, including: The cluster number setting submodule is used to set the cluster number K according to the new energy power evaluation requirements; The initial cluster center generation submodule is used to randomly generate multiple groups of initial cluster centers using a one-dimensional search method; The clustering submodule is used to cluster the normalized historical power generation data of new energy power stations using the initial clustering centers to obtain k clustering centers and k+1 power generation capacity levels.

10. The system according to claim 7, characterized in that Evaluation modules include: The evaluation background board drawing submodule is used to draw the evaluation background boards of different regions and seasons based on the normalized data clustering results and power generation capacity levels of new energy power generation in different regions, seasons and time periods, with time as the horizontal axis and 0-1 as the vertical axis.

Images