A distributed photovoltaic cluster power prediction method and device

Abstract

The present application relates to a kind of distributed photovoltaic cluster power prediction method and device, comprising: in each meteorological grid point in the total radiation forecast value vector of the region of distributed photovoltaic cluster in each historical day in given time, the similar vector of the total radiation forecast value vector of each meteorological grid point in the region of distributed photovoltaic cluster in given time in prediction day is selected;The actual total active power of distributed photovoltaic cluster corresponding to the similar vector is obtained;The actual total active power of distributed photovoltaic cluster corresponding to the similar vector is used to determine the total active power of distributed photovoltaic cluster in given time in prediction day;This method realizes distributed photovoltaic cluster power prediction by similarity degree sorting and median extraction, relative to traditional statistical upscaled method, this method has higher calculation stability and accuracy.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation prediction, specifically to a method and apparatus for predicting the power of a distributed photovoltaic cluster. Background Technology

[0002] In recent years, national photovoltaic development policies have gradually shifted towards distributed photovoltaic (PV) power generation, leading to a gradual expansion of its scale. Compared to centralized PV power plants, distributed PV is characterized by its numerous locations, wide distribution, and smaller individual unit capacity. Accurate prediction of PV power generation is crucial for grid dispatch planning and the optimized operation of PV power plants.

[0003] Traditional statistical upscaling methods for predicting cluster power are limited by the monitoring level and data quality of distributed photovoltaic systems, making traditional centralized photovoltaic power prediction methods less applicable. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of this invention is:

[0005] To address the problem of poor applicability of traditional statistical upscaling cluster power prediction methods, this invention provides a distributed photovoltaic cluster power prediction method with low data dependence, stability, reliability, and wide applicability.

[0006] The objective of this invention is achieved through the following technical solution:

[0007] This invention provides a method for predicting the power of a distributed photovoltaic cluster, the improvement of which is that the method includes:

[0008] In the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day, select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the forecast day.

[0009] Obtain the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector;

[0010] The total active power of the distributed photovoltaic cluster at a given moment in the predicted day is determined by using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0011] Preferably, the step of selecting a similar vector from the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day, and the vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the prediction day, includes:

[0012] Calculate the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day;

[0013] Select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

[0014] Furthermore, the calculation of the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day includes:

[0015] The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day. n The similarity d between n :

[0016]

[0017] Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs, and x i Let x be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. in Let X = (x1, x2, ..., xn) be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​of all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the nth historical day. i ...x m ), X n =(x 1n ,x 2n ...x in ...x mn ).

[0018] Preferably, determining the total active power of the distributed photovoltaic cluster at a given moment in the predicted day using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes:

[0019] The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day.

[0020] Furthermore, the process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes:

[0021] The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence;

[0022] If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0023] Based on the same inventive concept, the present invention also provides a distributed photovoltaic cluster power prediction device, wherein the improvement is that the device includes:

[0024] The selection module is used to select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day.

[0025] The acquisition module is used to acquire the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector;

[0026] The determination module is used to determine the total active power of the distributed photovoltaic cluster at a given moment in the predicted day by using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0027] Furthermore, the selection module includes:

[0028] The calculation unit is used to calculate the similarity between the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day.

[0029] The selection unit is used to select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

[0030] Furthermore, the computing unit is specifically used for:

[0031] The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day.n The similarity d between n :

[0032]

[0033] Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs, and x i Let x be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. in Let X = (x1, x2, ..., xn) be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​of all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the nth historical day. i ...x m ), X n =(x 1n ,x 2n ...x in ...x mn ).

[0034] Furthermore, the determining module is specifically used for:

[0035] The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day.

[0036] Furthermore, the process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes:

[0037] The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence;

[0038] If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] The technical solution provided by this invention first selects a similar vector to the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day. Second, it obtains the actual total active power of the distributed photovoltaic cluster corresponding to the similar vector. Finally, it determines the total active power of the distributed photovoltaic cluster at a given time in the forecast day using the actual total active power of the distributed photovoltaic cluster corresponding to the similar vector. This solution achieves a comprehensive description of the meteorological forecast characteristics of distributed photovoltaic clusters. The dataset relies solely on numerical weather prediction data and total active power data of distributed photovoltaic clusters, resulting in high data quality and reliability. This provides a reliable data foundation for prediction modeling. Compared to traditional statistical upscaling methods, this solution exhibits higher computational stability and accuracy. Attached Figure Description

[0041] Figure 1 This is a flowchart of the distributed photovoltaic cluster power prediction method in an embodiment of the present invention;

[0042] Figure 2 This is a graph showing the predicted power of the distributed photovoltaic cluster from June 30, 2020 to July 1, 2020, in an embodiment of the present invention.

[0043] Figure 3 This is a flowchart of the distributed photovoltaic cluster power prediction device in an embodiment of the present invention. Detailed Implementation

[0044] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] Because existing statistical upscaling cluster power prediction methods have poor applicability, this invention proposes a stable and reliable method and apparatus for predicting distributed photovoltaic cluster power, such as... Figure 1 As shown, it includes:

[0047] 101. Select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day.

[0048] 102. Obtain the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector;

[0049] 103. The total active power of the distributed photovoltaic cluster at a given moment in the predicted day is determined by using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0050] To achieve 101, the optimal embodiment provided by the present invention specifically includes:

[0051] Calculate the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day;

[0052] Select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

[0053] In the optimal embodiment provided by the present invention, the preset value k can be 6, that is, the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the 6 historical days with the highest similarity is selected.

[0054] Specifically, in the optimal embodiment provided by the present invention, the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the historical day can be calculated using the following method:

[0055] The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day. n The similarity d between n :

[0056]

[0057] Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs, and x i Let x be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. in Let X = (x1, x2, ..., xn) be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​of all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the nth historical day. i ...xm ), X n =(x 1n ,x 2n ...x in ...x mn ).

[0058] Furthermore, in the optimal embodiment provided by the present invention, the total number m of meteorological grid points in the region to which the distributed photovoltaic cluster belongs in the above similarity formula is 72. The total radiation forecast value vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given moment during the predicted day is divided into two vectors X1 and X2, respectively, at a time interval of 15 minutes. n , obtain X and X n The number of sets is 96, where sets X and X' are respectively... n The number of data entries is 120.

[0059] In the optimal embodiment provided by the present invention, the preset value can be 6, that is, the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the 6 historical days with the highest similarity is selected, and the 6 actual active power data corresponding to the total radiation forecast value of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the 6 historical days with the highest similarity are obtained as (1000MW, 1200MW, 1500MW, 1500MW, 1700MW, 1750MW).

[0060] To achieve 103, the optimal embodiment provided by the present invention specifically includes:

[0061] The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day.

[0062] The process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes:

[0063] The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence;

[0064] If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0065] In the optimal embodiment provided by the present invention, the preset value k can be 6, that is, the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the 6 historical days with the highest similarity is selected. Therefore, the number of total radiation forecast value vectors of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the historical days with the highest similarity is 6. The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similar vector is the average of the two middle data.

[0066] In the preferred embodiment provided by the present invention, Figure 2 The results show the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the forecast day, calculated according to the above steps, and the total radiation actual vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the forecast day.

[0067] In the preferred embodiment provided by the present invention, Figure 2 In the diagram, the solid line represents the measured power; the dashed line represents the predicted power.

[0068] Based on the same inventive concept, this invention also provides a distributed photovoltaic cluster power prediction device, the optimal embodiment of which is as follows: Figure 3 As shown, it specifically includes:

[0069] The selection module is used to select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day.

[0070] The acquisition module is used to acquire the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector;

[0071] The determination module is used to determine the total active power of the distributed photovoltaic cluster at a given moment in the predicted day by using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0072] In the preferred embodiment provided by the present invention, the selection module includes:

[0073] The calculation unit is used to calculate the similarity between the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast value vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day.

[0074] The selection unit is used to select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

[0075] Furthermore, the computing unit is specifically used for:

[0076] The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day. n The similarity d between n :

[0077]

[0078] Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs, and x i Let x be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. in Let X = (x1, x2, ..., xn) be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​of all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the nth historical day. i ...x m ), X n =(x 1n ,x 2n ...x in ...x mn ).

[0079] In the preferred embodiment provided by this invention, the determining module is specifically used for:

[0080] The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day.

[0081] Furthermore, the process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes:

[0082] The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence;

[0083] If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

[0084] 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.

[0085] 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.

[0086] 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 1 The function specified in one or more boxes.

[0087] 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.

[0088] 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 predicting the power of a distributed photovoltaic cluster, characterized in that, The method includes: In the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day, select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in the forecast day. Obtain the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; The total active power of the distributed photovoltaic cluster at a given moment in the midday is determined by using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; The step of determining the total active power of the distributed photovoltaic cluster at a given moment in the daytime using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes: The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day. The process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes: The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence; If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

2. The method as described in claim 1, characterized in that, The step of selecting similar vectors from the total radiation forecast vectors of meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day, including: Calculate the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day; Select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

3. The method as described in claim 2, characterized in that, The calculation of the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time of the predicted midday and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time of the historical midday includes: The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day. n The similarity d between n : Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs. The total radiation forecast value is the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. Let be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the n-th historical day. , .

4. A distributed photovoltaic cluster power prediction device, characterized in that, The device includes: The selection module is used to select the similar vector of the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time in each historical day. The acquisition module is used to acquire the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; The determination module is used to determine the total active power of the distributed photovoltaic cluster at a given moment in the midday prediction using the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector. The determining module is specifically used for: The median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector is used as the predicted total active power of the distributed photovoltaic cluster at a given time during the day. The process of obtaining the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector includes: The actual total active power of the distributed photovoltaic clusters corresponding to the similarity vectors is sorted in descending order to obtain the power sorting sequence; If the number of data in the power sorting sequence is odd, then the data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector; otherwise, the average of the two data in the middle of the power sorting sequence is taken as the median of the actual total active power of the distributed photovoltaic cluster corresponding to the similarity vector.

5. The apparatus as described in claim 4, characterized in that, The selection module includes: The calculation unit is used to calculate the similarity between the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day and the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during each historical day. The selection unit is used to select the total radiation forecast vector of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time from the k historical days with the highest similarity, where k is a preset value.

6. The apparatus as claimed in claim 5, characterized in that, The computing unit is specifically used for: The total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day is calculated using the following formula, along with the total radiation forecast vector X of each meteorological grid point in the region to which the distributed photovoltaic cluster belongs at a given time during the nth historical day. n The similarity d between n : Where m is the total number of meteorological grid points in the region to which the distributed photovoltaic cluster belongs. The total radiation forecast value is the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time during the predicted day. Let be the total radiation forecast value of the i-th meteorological grid point in the vector of total radiation forecast values ​​for all meteorological grid points in the region to which the distributed photovoltaic cluster belongs at a given time in the nth historical day. , .

Images