Regional distributed photovoltaic power prediction method and system based on DBSCAN

Abstract

The invention discloses a regional distributed photovoltaic power prediction method and system based on DBSCAN, and the method comprises the steps: obtaining the real-time power data and power station meteorological data of photovoltaic power stations in a region, carrying out the clustering analysis of the single-day maximum photovoltaic power of the photovoltaic power stations, and dividing the photovoltaic power stations with the same characteristics into a same cluster; if the photovoltaic power station in any cluster has meteorological data, selecting the meteorological data of the photovoltaic power station with the maximum generation power to represent the whole cluster; if not, obtaining the meteorological data by interpolating the meteorological data of the adjacent stations; a prediction model is trained according to the formed cluster and the meteorological information, the prediction model takes the original power accumulated value and the meteorological condition of the photovoltaic power station in the cluster as the input, and the output is predicted power; and combining the trained prediction model with weather forecast information to complete photovoltaic power prediction in the region. When the method is used for prediction, the number of groups does not need to be specified in advance, the design is simplified, and power prediction of the whole area is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of energy forecasting and machine learning, and in particular to a regional distributed photovoltaic power forecasting method and system based on DBSCAN. Background Art

[0002] In the field of photovoltaic power generation, power prediction of photovoltaic power stations is crucial to achieve efficient energy management and optimize grid operation. Current power prediction methods involve multiple influencing factors such as wind speed, wind direction, temperature, humidity, air pressure and light intensity, among which the BP neural network method has been widely used. However, if there are many photovoltaic power stations in a region, directly using the data of all stations for prediction will lead to excessive calculation and increased model complexity.

[0003] In addition, when there is a large amount of photovoltaic power generation in a certain area, some of the large installed capacity is directly connected to the grid through medium voltage (10kV). This part has a complete environmental collection system, which can collect relevant data such as wind speed, wind direction, temperature, humidity, air pressure, light intensity, real-time power, etc., and has the conditions for power prediction; and most of the installed capacity is not large, and it is directly connected to the grid through low voltage (380V). This part can only collect real-time power and does not have the conditions for power prediction. Since low-voltage grid-connected photovoltaic power stations do not have the conditions for power prediction, it is necessary to find a solution to the problem and predict this part in order to achieve regional prediction. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a regional distributed photovoltaic power prediction method and system based on DBSCAN to solve the problems mentioned in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a regional distributed photovoltaic power prediction method based on DBSCAN, comprising: acquiring real-time power data and meteorological data of photovoltaic power stations in a region, and performing cluster analysis on the maximum photovoltaic power of the photovoltaic power station in a single day, and dividing photovoltaic power stations with the same characteristics into the same cluster;

[0009] If the meteorological data of the photovoltaic power station in any of the clusters exists, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; if it does not exist, the meteorological data is obtained by interpolation of the meteorological data of the adjacent stations;

[0010] A prediction model is trained according to the formed cluster and meteorological information, wherein the prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and outputs the predicted power;

[0011] The trained prediction model is combined with weather forecast information to complete the photovoltaic power prediction in the area.

[0012] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, before clustering analysis is performed on the single-day maximum photovoltaic power of the photovoltaic power station, it also includes: standardizing and normalizing the real-time power data of the photovoltaic power stations in the acquisition area to obtain N power curves, wherein the power curves are time-regularized and all curves are data of the same time period.

[0013] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, wherein: clustering analysis is performed on the daily maximum photovoltaic power of the photovoltaic power station, and photovoltaic power stations with the same characteristics are divided into the same cluster, including: clustering analysis is performed on the daily maximum photovoltaic power of N pre-processed photovoltaic power stations using the DBSCAN method, and photovoltaic power stations with the same characteristics are classified into one category;

[0014] The maximum photovoltaic power sample set is expressed as:

[0015] D={P max,1 , P max,2 , ..., P max,N}

[0016] Among them, P max,N is the daily maximum photovoltaic power of the Nth photovoltaic power station;

[0017] The sample distance metric is expressed as:

[0018]

[0019] Among them, P max,i,d is the maximum power generation of the ith PV plant on the dth day, P max,j,d is the maximum power generation of the j-th PV power station on the d-th day.

[0020] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, it also includes: expanding any core point in the maximum photovoltaic power sample set; traversing all core points in the neighborhood of the core point, obtaining all density-connected data points starting from the core point, until there are no expanded data points, and the boundary nodes of the clusters formed by clustering are non-core data points;

[0021] Similarly, the unclustered data points in the maximum photovoltaic power sample set are traversed until there are no new core points in the data set, and the data points in the data set that are not included in any cluster are constituted as outliers.

[0022] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, wherein: training the prediction model according to the formed cluster and meteorological information includes: using the neural network BP algorithm to train the model on the obtained data, taking the original power accumulation value of each photovoltaic power station in the cluster and the meteorological conditions as the features of the model input, and accumulating the predicted power as the target output of the model; the meteorological conditions include wind speed, wind direction, speed, temperature, humidity, air pressure, and light intensity;

[0023] When the mean square error between the predicted power and the actual power is the smallest, the iteration is stopped and M prediction models are output.

[0024] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, it also includes: the neural network BP algorithm is a single hidden layer BP neural network, including an input layer, a hidden layer and an output layer, the number of neurons in the input layer is consistent with the input characteristics of the data, and the number of output neurons is 1;

[0025] The calculation formula for selecting the number of hidden layer neurons is:

[0026]

[0027] Among them, l is the number of neurons in the hidden layer, m is the number of neurons in the input layer, n is the number of neurons in the output layer, and h is the adjustment constant, which is generally 1 to 10;

[0028] The Sigmoid function is selected as the activation function of the neurons in the hidden layer and output layer of the BP neural network, and the cross-validation method is used to train and verify the model.

[0029] As a preferred solution of the regional distributed photovoltaic power prediction method based on DBSCAN described in the present invention, the trained prediction model is combined with weather forecast information to complete the photovoltaic power prediction in the region, including: if meteorological data exists for the photovoltaic power station in any cluster, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; if not, the meteorological data is obtained by interpolation of meteorological data from adjacent sites; and the power of all power stations in the current area is predicted according to the weather forecast information through the generated prediction model.

[0030] In a second aspect, the present invention provides a regional distributed photovoltaic power prediction system based on DBSCAN, comprising:

[0031] The cluster division module is used to obtain the real-time power data and meteorological data of the photovoltaic power stations in the region, and to perform cluster analysis on the maximum photovoltaic power of the photovoltaic power stations in a single day, and to divide the photovoltaic power stations with the same characteristics into the same cluster;

[0032] A meteorological data processing module, for selecting the meteorological data of the photovoltaic power station with the largest power generation power to represent the entire cluster if the meteorological data of the photovoltaic power station in any of the clusters exists; if the meteorological data does not exist, the meteorological data is obtained by interpolation of the meteorological data of the adjacent stations;

[0033] A model training module, used for training a prediction model according to the formed cluster and meteorological information, wherein the prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and outputs the predicted power;

[0034] The prediction module is used to combine the trained prediction model with weather forecast information to complete the photovoltaic power prediction in the area.

[0035] In a third aspect, the present invention provides a computing device, comprising:

[0036] Memory and processor;

[0037] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the regional distributed photovoltaic power prediction method based on DBSCAN are implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the regional distributed photovoltaic power prediction method based on DBSCAN.

[0039] Compared with the prior art, the present invention has the following beneficial effects: the present invention divides N photovoltaic power stations in a region into M clusters by using the DBSCAN method, and does not need to specify the number of groups in advance, thus simplifying the design; and realizes power prediction for the entire region. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0041] Figure 1 The present invention is a flowchart of a regional distributed photovoltaic power prediction method and system based on DBSCAN according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0045] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0046] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0047] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0048] Example 1

[0049] Reference Figure 1 , is an embodiment of the present invention, which provides a regional distributed photovoltaic power prediction method based on DBSCAN, comprising:

[0050] S100: Acquire real-time power data and meteorological data of photovoltaic power stations in the region, perform cluster analysis on the maximum photovoltaic power of photovoltaic power stations in a single day, and divide photovoltaic power stations with the same characteristics into the same cluster;

[0051] S200: If meteorological data exists for any photovoltaic power station in any cluster, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; if not, the meteorological data is obtained by interpolation of the meteorological data of the adjacent stations;

[0052] S300: training a prediction model based on the formed cluster and meteorological information, the prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and outputs the predicted power;

[0053] S400: Combine the trained prediction model with weather forecast information to complete the photovoltaic power prediction in the area.

[0054] It should be noted that the real-time power and other related data of all photovoltaic power stations are collected. If a photovoltaic power station has meteorological data, the corresponding meteorological data will be obtained at the same time, including but not limited to the wind speed, wind direction, temperature, humidity, air pressure, light intensity, etc. of the photovoltaic power station.

[0055] It should also be noted that the preprocessing of the collected data includes data cleaning: processing of outliers, missing values ​​and noise values.

[0056] In an embodiment of the present application, before clustering analysis of the maximum daily photovoltaic power of a photovoltaic power station is performed, it also includes: standardizing and normalizing the real-time power data of photovoltaic power stations in the acquisition area to obtain N power curves, wherein the power curves are time-regularized and all curves are data for the same time period.

[0057] For example, in one year, a power curve is formed according to a collection point every 15 minutes. In one year, 365*24*4=35040 points are accumulated. Each point of real-time power is recorded as: P ij , where i represents the i-th medium voltage curve, ranging from 1 to M, and j ranges from 1 to 35040.

[0058] Specifically, the power of each power station is normalized by calculating the average value of the largest m points (for example, m is 1000) of each curve and dividing the real-time power by the average value. Each normalized point is denoted as P ij ′, if the normalized value exceeds 1, it is calculated as 1. At this time, the normalized value becomes the [0, 1] interval.

[0059] Specifically, the normalized data is analyzed on a daily basis, and the 35,040 points are divided into 365 arrays with 96 points per day. For each day's data, the average of the four largest data points is calculated. In this way, each curve corresponds to 365 average values. The maximum power of the ith PV power station (i ranges from 1 to N) on the dth day (d ranges from 0 to 365) is defined as P max,i,d .

[0060] It should be noted that the maximum daily photovoltaic power of photovoltaic power plants with the same characteristics should be consistent. In the embodiment of the present application, cluster analysis is performed on the maximum daily photovoltaic power of photovoltaic power stations, and photovoltaic power stations with the same characteristics are divided into the same cluster, including: clustering the maximum daily photovoltaic power of N photovoltaic power stations after preprocessing using the DBSCAN method, and classifying photovoltaic power stations with the same characteristics into one category;

[0061] The maximum photovoltaic power sample set is expressed as:

[0062] D={P max,1 , P max,2 , ..., P max,N}

[0063] Among them, P max,N is the daily maximum photovoltaic power of the Nth photovoltaic power station;

[0064] The sample distance metric is expressed as:

[0065]

[0066] Among them, Pmax,i,d is the maximum power generation of the ith PV plant on the dth day, P max,j,d is the maximum power generation of the j-th PV power station on the d-th day.

[0067] It should be noted that the neighborhood parameters are: the object radius E, which can be set by computer input; the minimum number MinPts, which can be set by computer input, and can be set to 3 in the embodiment of the present application.

[0068] In the embodiment of the present application, it also includes: expanding any core point in the maximum photovoltaic power sample set; since the boundary point cannot be expanded, traversing all core points in the neighborhood of the core point, obtaining all density-connected data points starting from the core point, until there are no expanded data points, and the boundary nodes of the clusters formed by clustering are non-core data points;

[0069] Similarly, the unclustered data points in the maximum photovoltaic power sample set are traversed until there are no new core points in the data set, and the data points in the data set that are not included in any cluster are constituted as outliers.

[0070] In the embodiment of the present application, training a prediction model based on the formed cluster and meteorological information includes: using a neural network BP algorithm to train the model on the acquired data, taking the original power accumulation value of each photovoltaic power station in the cluster and the meteorological conditions as the features of the model input, and accumulating the predicted power as the target output of the model; the meteorological conditions include wind speed, wind direction, speed, temperature, humidity, air pressure, and light intensity;

[0071] When the mean square error between the predicted power and the actual power is the smallest, the iteration is stopped and M prediction models are output.

[0072] In the embodiment of the present application, it also includes: the neural network BP algorithm is a single hidden layer BP neural network, including an input layer, a hidden layer and an output layer, the number of neurons in the input layer is consistent with the input characteristics of the data, and the number of output neurons is 1;

[0073] The calculation formula for selecting the number of hidden layer neurons is:

[0074]

[0075] Among them, l is the number of neurons in the hidden layer, m is the number of neurons in the input layer, n is the number of neurons in the output layer, and h is the adjustment constant, which is generally 1 to 10;

[0076] The Sigmoid function is selected as the activation function of the neurons in the hidden layer and output layer of the BP neural network, and the cross-validation method is used to train and verify the model.

[0077] In an embodiment of the present application, the trained prediction model is combined with weather forecast information to complete the photovoltaic power prediction in the area, including: if meteorological data exists for any photovoltaic power station in a cluster, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; if it does not exist, the meteorological data is obtained by interpolation of adjacent stations; and the power of all power stations in the current area is predicted based on the weather forecast information using the generated prediction model.

[0078] It should be noted that the present invention divides N photovoltaic power stations in a region into M clusters by using the DBSCAN method, without specifying the number of groups in advance, thus simplifying the design; and ultimately achieving power prediction for the entire region.

[0079] Example 2

[0080] In an exemplary embodiment, a regional distributed photovoltaic power prediction system based on DBSCAN is also provided in this embodiment, including:

[0081] The cluster division module is used to obtain the real-time power data and meteorological data of the photovoltaic power stations in the region, and to perform cluster analysis on the maximum photovoltaic power of the photovoltaic power stations in a single day, and to divide the photovoltaic power stations with the same characteristics into the same cluster;

[0082] The meteorological data processing module is used to select the meteorological data of the photovoltaic power station with the largest power generation power to represent the entire cluster if there is meteorological data for the photovoltaic power station in any cluster; if there is no meteorological data, the meteorological data of the photovoltaic power station with the largest power generation power is obtained by interpolation of meteorological data of adjacent stations;

[0083] The model training module is used to train the prediction model based on the formed cluster and meteorological information. The prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and the output is the predicted power;

[0084] The prediction module is used to combine the trained prediction model with weather forecast information to complete the photovoltaic power prediction in the area.

[0085] Furthermore, the regional distributed photovoltaic power prediction system based on DBSCAN also includes:

[0086] The preprocessing module is used to standardize and normalize the real-time power data of the photovoltaic power station in the acquisition area to obtain N power curves, where the power curves are time-regularized and all curves are data in the same time period.

[0087] In an optional embodiment, the regional distributed photovoltaic power prediction system based on DBSCAN may perform the following steps:

[0088] The real-time power data of the photovoltaic power stations in the acquisition area are standardized and normalized to obtain N power curves, among which the power curves are time-regularized and all curves are data in the same time period.

[0089] The DBSCAN method is used to perform cluster analysis on the daily maximum photovoltaic power of N photovoltaic power stations after preprocessing, and photovoltaic power stations with the same characteristics are classified into one category;

[0090] The maximum photovoltaic power sample set is expressed as:

[0091] D={P max,1 , P max,2 , ..., P max,N}

[0092] Among them, P max,N is the daily maximum photovoltaic power of the Nth photovoltaic power station;

[0093] The sample distance metric is expressed as:

[0094]

[0095] Among them, P max,i,d is the maximum power generation of the ith PV plant on the dth day, P max,j,d is the maximum power generation of the j-th PV power station on the d-th day.

[0096] Expand any core point in the maximum photovoltaic power sample set; traverse all core points in the neighborhood of the core point, and obtain all density-connected data points starting from the core point until there are no expanded data points, and the boundary nodes of the clusters formed by clustering are non-core data points;

[0097] Similarly, the unclustered data points in the maximum photovoltaic power sample set are traversed until there are no new core points in the data set, and the data points in the data set that are not included in any cluster are constituted as outliers.

[0098] The neural network BP algorithm is used to train the model with the acquired data. The original power accumulation value of each photovoltaic power station in the cluster and the meteorological conditions are used as the input features of the model. The accumulated predicted power is used as the target output of the model. The meteorological conditions include wind speed, wind direction, speed, temperature, humidity, air pressure, and light intensity.

[0099] When the mean square error between the predicted power and the actual power is the smallest, the iteration is stopped and M prediction models are output.

[0100] The neural network BP algorithm is a single hidden layer BP neural network, including input layer, hidden layer and output layer. The number of neurons in the input layer is consistent with the input characteristics of the data, and the number of output neurons is 1.

[0101] The calculation formula for selecting the number of hidden layer neurons is:

[0102]

[0103] Among them, l is the number of neurons in the hidden layer, m is the number of neurons in the input layer, n is the number of neurons in the output layer, and h is the adjustment constant, which is generally 1 to 10;

[0104] The Sigmoid function is selected as the activation function of the neurons in the hidden layer and output layer of the BP neural network, and the cross-validation method is used to train and verify the model.

[0105] If meteorological data exists for any PV power station in a cluster, the meteorological data of the PV power station with the largest power generation capacity is selected to represent the entire cluster. If it does not exist, the meteorological data is obtained by interpolation of the meteorological data of adjacent stations. The generated prediction model is used to predict the power of all power stations in the current area based on weather forecast information.

[0106] This embodiment further provides a computing device, which is applicable to the regional distributed photovoltaic power prediction method based on DBSCAN, including:

[0107] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the regional distributed photovoltaic power prediction method based on DBSCAN as proposed in the above embodiment.

[0108] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the regional distributed photovoltaic power prediction method based on DBSCAN as proposed in the above embodiment is implemented.

[0109] The storage medium proposed in this embodiment and the method for implementing regional distributed photovoltaic power prediction based on DBSCAN proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0110] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0111] 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A regional distributed photovoltaic power prediction method based on DBSCAN, characterized in that: include: Acquire the real-time power data and meteorological data of the photovoltaic power stations in the region, perform cluster analysis on the maximum photovoltaic power of the photovoltaic power stations in a single day, and divide the photovoltaic power stations with the same characteristics into the same cluster; If the photovoltaic power station in any of the clusters has meteorological data, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; If it does not exist, it is obtained by interpolating the meteorological data of the adjacent stations; A prediction model is trained according to the formed cluster and meteorological information, wherein the prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and outputs the predicted power; The trained prediction model is combined with weather forecast information to complete the photovoltaic power prediction in the area.

2. The regional distributed photovoltaic power prediction method based on DBSCAN according to claim 1, characterized in that: Before clustering analysis of the maximum daily photovoltaic power of the photovoltaic power station, it also includes: standardizing and normalizing the real-time power data of the photovoltaic power station in the acquisition area to obtain N power curves, wherein the power curves are time-regularized and all curves are data of the same time period.

3. The regional distributed photovoltaic power prediction method based on DBSCAN as claimed in claim 2, characterized in that: Performing cluster analysis on the daily maximum photovoltaic power of the photovoltaic power station, and classifying photovoltaic power stations with the same characteristics into the same cluster includes: performing cluster analysis on the daily maximum photovoltaic power of the N photovoltaic power stations after preprocessing by using the DBSCAN method, and classifying photovoltaic power stations with the same characteristics into one category; The maximum photovoltaic power sample set is expressed as: D={P max,1 ,P max,2 ,...,P max,N } Among them, P max,N is the daily maximum photovoltaic power of the Nth photovoltaic power station; The sample distance metric is expressed as: Among them, P max,i,d is the maximum power generation of the ith PV plant on the dth day, P max,j,d is the maximum power generation of the j-th PV power station on the d-th day.

4. The regional distributed photovoltaic power prediction method based on DBSCAN as claimed in claim 3, characterized in that: It also includes: expanding any core point in the maximum photovoltaic power sample set; traversing all core points in the neighborhood of the core point, obtaining all density-connected data points starting from the core point, until there are no expanded data points, and the boundary nodes of the clusters formed by clustering are non-core data points; Similarly, the unclustered data points in the maximum photovoltaic power sample set are traversed until there are no new core points in the data set, and the data points in the data set that are not included in any cluster are constituted as outliers.

5. The regional distributed photovoltaic power prediction method based on DBSCAN as claimed in claim 4, characterized in that: Training a prediction model based on the formed cluster and meteorological information includes: using a neural network BP algorithm to train the model on the acquired data, taking the original power accumulation value of each photovoltaic power station in the cluster and the meteorological conditions as the features of the model input, and accumulating the predicted power as the target output of the model; the meteorological conditions include wind speed, wind direction, speed, temperature, humidity, air pressure, and light intensity; When the mean square error between the predicted power and the actual power is the smallest, the iteration is stopped and M prediction models are output.

6. The regional distributed photovoltaic power prediction method based on DBSCAN as claimed in claim 5, characterized in that: Also includes: The neural network BP algorithm is a single hidden layer BP neural network, including an input layer, a hidden layer and an output layer. The number of neurons in the input layer is consistent with the input characteristics of the data, and the number of output neurons is 1; The calculation formula for selecting the number of hidden layer neurons is: Among them, l is the number of neurons in the hidden layer, m is the number of neurons in the input layer, n is the number of neurons in the output layer, and h is the adjustment constant, which is generally 1 to 10; The Sigmoid function is selected as the activation function of the neurons in the hidden layer and output layer of the BP neural network, and the cross-validation method is used to train and verify the model.

7. The regional distributed photovoltaic power prediction method based on DBSCAN according to claim 6, characterized in that: The trained prediction model is combined with weather forecast information to complete the photovoltaic power prediction in the area, including: if meteorological data of the photovoltaic power station in any cluster exists, the meteorological data of the photovoltaic power station with the largest power generation power is selected to represent the entire cluster; if it does not exist, it is obtained by interpolating the meteorological data of the adjacent stations; and the power of all power stations in the current area is predicted according to the weather forecast information through the generated prediction model.

8. A regional distributed photovoltaic power prediction system based on DBSCAN, characterized in that: include: The cluster division module is used to obtain the real-time power data and meteorological data of the photovoltaic power stations in the region, and to perform cluster analysis on the maximum photovoltaic power of the photovoltaic power stations in a single day, and to divide the photovoltaic power stations with the same characteristics into the same cluster; A meteorological data processing module, for selecting the meteorological data of the photovoltaic power station with the largest power generation to represent the entire cluster if the photovoltaic power station in any of the clusters has meteorological data; If it does not exist, it is obtained by interpolating the meteorological data of the adjacent stations; A model training module, used for training a prediction model according to the formed cluster and meteorological information, wherein the prediction model takes the original power accumulation value of the photovoltaic power station in the cluster and the meteorological conditions as input, and outputs the predicted power; The prediction module is used to combine the trained prediction model with weather forecast information to complete the photovoltaic power prediction in the area.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the regional distributed photovoltaic power prediction method based on DBSCAN are implemented in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the regional distributed photovoltaic power prediction method based on DBSCAN as described in any one of claims 1 to 7.

Images