Distributed photovoltaic day-ahead scene prediction method, device, equipment and medium

Abstract

The invention discloses a distributed photovoltaic day-ahead scene prediction method and device, equipment and a medium. The method is characterized by comprising the following steps: acquiring first historical photovoltaic information of distributed photovoltaic and second historical photovoltaic information of centralized photovoltaic; determining a virtual power station corresponding to the distributed photovoltaic and power generation information of the virtual power station according to the first historical photovoltaic information and the second historical photovoltaic information; performing photovoltaic power generation prediction on the virtual power station through a pre-trained photovoltaic prediction model according to the power generation information of the virtual power station, and determining a day-ahead power curve; and collecting photovoltaic scene data of a virtual power station, performing prediction according to the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network, and determining target photovoltaic scene data. According to the invention, the day-ahead scene of each distributed photovoltaic can be predicted, the prediction quality of the day-ahead scene of the distributed photovoltaic is improved, and self-power prediction is effectively carried out in the distributed terminal corresponding to the distributed photovoltaic.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power generation, and in particular to a method, device, equipment and medium for day-ahead scene prediction of distributed photovoltaics. Background Art

[0002] With the continuous expansion of the power system and the vigorous development of renewable resources, a large number of distributed photovoltaics have been incorporated into the power system. However, since distributed photovoltaics are affected by factors such as weather conditions, geographical location and seasonal changes, the output of distributed photovoltaics has volatility and random characteristics, resulting in a large amount of data generated by distributed photovoltaics when connected to the power system. The power system is difficult to manage and process these large amounts of data. The edge cloud collaborative architecture can effectively alleviate the pressure of the power system on data management and processing. The edge cloud collaborative architecture requires that the edge nodes where distributed photovoltaics are located can effectively predict the power generation data of distributed photovoltaics, thereby reducing the data pressure generated by distributed photovoltaics. In the existing technology, it is difficult for the edge nodes of distributed photovoltaics to effectively predict the day-ahead scene data, resulting in the low accuracy of the prediction of the power generation power by the edge nodes of distributed photovoltaics. Summary of the invention

[0003] The present invention provides a distributed photovoltaic day-ahead scene prediction method, device, equipment and medium to solve the technical problem that the distributed photovoltaic day-ahead scene cannot be accurately predicted.

[0004] According to one aspect of the present invention, a method for day-ahead scene prediction of distributed photovoltaics is provided, comprising:

[0005] Acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics;

[0006] Determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information;

[0007] Predicting photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station by using a pre-trained photovoltaic prediction model to determine a day-ahead power curve;

[0008] Photovoltaic scene data of a virtual power station is collected, and prediction is performed based on the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine target photovoltaic scene data.

[0009] According to another aspect of the present invention, a distributed photovoltaic day-ahead scene prediction device is provided, comprising:

[0010] A data acquisition module, used to acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics;

[0011] A virtual power station combination module, used to determine the virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information;

[0012] A power curve prediction module, used to predict photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station through a pre-trained photovoltaic prediction model, and determine a day-ahead power curve;

[0013] The photovoltaic day-ahead scene prediction module is used to collect photovoltaic scene data of the virtual power station, and predict the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine the target photovoltaic scene data.

[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the day-ahead scene prediction method for distributed photovoltaics described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the distributed photovoltaic day-ahead scene prediction method described in any embodiment of the present invention when executed.

[0019] The technical solution of the embodiment of the present invention, by acquiring the first historical photovoltaic information of distributed photovoltaic and the second historical photovoltaic information of centralized photovoltaic, the historical photovoltaic information is used as the basis for prediction, which can make the prediction result closer to reality and improve the rationality of the prediction result; according to the first historical photovoltaic information and the second historical photovoltaic information, the virtual power station and the virtual power station power generation information corresponding to the distributed photovoltaic are determined, and the utilization rate and flexibility of data calculation are improved through the data sharing of distributed photovoltaic and centralized photovoltaic, which can effectively reduce the calculation complexity and indirectly improve the efficiency of prediction; through the pre-trained photovoltaic prediction model, the photovoltaic power generation of the virtual power station is predicted according to the virtual power station power generation information, and the day-ahead power curve is determined, which can accurately capture each dependency in the historical information, and improve the prediction accuracy through the powerful feature extraction ability of the long-term and short-term neural network; the photovoltaic scene data of the virtual power station is collected, and the photovoltaic scene data and the day-ahead power curve of the virtual power station are predicted through the conditional generative adversarial network to determine the target photovoltaic scene data. The conditional generative adversarial network can effectively improve the quality of the day-ahead scene, and solve the technical problem that the distributed photovoltaic day-ahead scene cannot be accurately predicted in the prior art. It is possible to predict the day-ahead scenarios of each distributed photovoltaic system, improve the prediction quality of the day-ahead scenarios of distributed photovoltaic systems, and accurately predict the power in the distributed terminals corresponding to the distributed photovoltaic systems.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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 work.

[0022] Figure 1 A flowchart of a method for day-ahead scene prediction of distributed photovoltaics is provided for an embodiment of the present invention;

[0023] Figure 2 A flowchart of another distributed photovoltaic day-ahead scenario prediction method provided by an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of the structure of a distributed photovoltaic day-ahead scene prediction device provided by an embodiment of the present invention;

[0025] Figure 4A schematic structural diagram of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Figure 1 A flowchart of a distributed photovoltaic day-ahead scene prediction method is provided for an embodiment of the present invention. This embodiment is applicable to the case where a distributed photovoltaic day-ahead scene is generated when the distributed photovoltaic power generation is predicted after the distributed photovoltaic is connected to the grid. The method can be executed by a distributed photovoltaic day-ahead scene prediction device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0029] S110, obtaining first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics.

[0030] Among them, distributed photovoltaic can be photovoltaic power generation equipment built by grid-connected users themselves. The number of distributed photovoltaic can be 1 or more; for example, distributed photovoltaic can be a photovoltaic power generation project on the roof of an urban building or a rural building. Centralized photovoltaic can be a large-scale photovoltaic power generation equipment built in a centralized manner. Centralized photovoltaic is usually in areas with abundant light resources, and large-scale photovoltaic panel arrays can be built; the number of centralized photovoltaic can be 1 or more; for example, centralized photovoltaic can be a large photovoltaic panel array built in a large industrial park.

[0031] Among them, the first historical photovoltaic information can be the distributed photovoltaic power generation information recorded every day in history; the first historical photovoltaic information is mainly composed of the weather information and power generation information of the day, the first historical photovoltaic information includes a first power sequence and a first meteorological data, the first power sequence can be a series composed of the distributed photovoltaic power generation power values ​​of each day and each time period in history arranged in chronological order, and the first meteorological data can be the weather data corresponding to the distributed photovoltaic power generation power value, such as light intensity, ambient temperature and weather conditions.

[0032] Among them, the second historical photovoltaic information can be a record of centralized photovoltaic power generation information for each day in history; the second historical photovoltaic information is mainly composed of the weather information and power generation information of the day, the second historical photovoltaic information includes a second power sequence and second meteorological data, the second power sequence can be a series composed of the centralized photovoltaic power generation power values ​​of each day and each time period in history arranged in chronological order, and the second meteorological data can be the weather data corresponding to the centralized photovoltaic power generation power values.

[0033] Optionally, in an embodiment of the present invention, distributed photovoltaics and centralized photovoltaics are respectively used as edge computing nodes of the power system. Distributed photovoltaics and centralized photovoltaics can automatically collect and record photovoltaic power generation power values ​​and weather data in historical time periods. Distributed photovoltaics combine photovoltaic power generation power values ​​and weather data into a first power sequence and first meteorological data according to time sequence to obtain first historical photovoltaic information; centralized photovoltaics combine photovoltaic power generation power values ​​and weather data into a second power sequence and second meteorological data according to time sequence to obtain second historical photovoltaic information.

[0034] Specifically, for each distributed photovoltaic, the first historical photovoltaic information of each distributed photovoltaic is obtained, and for the centralized photovoltaic, the second historical photovoltaic information of each centralized photovoltaic is obtained.

[0035] S120. Determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information.

[0036] Among them, the virtual power station can be a photovoltaic power node formed by the mutual communication and data exchange between distributed photovoltaic and centralized photovoltaic. It should be noted that the virtual power station is constructed by allocating each distributed photovoltaic to each centralized photovoltaic, with each centralized photovoltaic as the center and each distributed photovoltaic.

[0037] The virtual power station power generation information may be the power generation power sequence and weather data of the virtual power station. It should be noted that the virtual power station power generation information is composed of the centralized photovoltaic and distributed photovoltaic of each virtual power station, and the second power sequence and second meteorological data of the centralized photovoltaic of each virtual power station are combined with the first power sequence and first meteorological data of the distributed photovoltaic of the virtual power station to obtain the virtual power station power generation information.

[0038] Optionally, each distributed photovoltaic and each centralized photovoltaic are respectively combined according to the first historical photovoltaic information and the second historical photovoltaic information to form a virtual power station corresponding to each distributed photovoltaic, and the first historical photovoltaic information corresponding to the distributed photovoltaics constituting the virtual power station and the second historical photovoltaic information corresponding to the centralized photovoltaics are combined to obtain the virtual power station power generation information.

[0039] Optionally, in another optional embodiment of the present invention, determining the virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information includes:

[0040] According to the first historical photovoltaic information and the second historical photovoltaic information, a target cluster photovoltaic corresponding to each distributed photovoltaic is determined in the centralized photovoltaic; the distributed photovoltaics are clustered into the target cluster photovoltaic, and the virtual power station and the power generation information of the virtual power station are determined.

[0041] The target cluster PV can be the centralized PV to which the distributed PV is assigned. It should be noted that for a distributed PV, it is necessary to conduct spatial correlation with each centralized PV, and select the centralized PV with the largest spatial correlation as the target cluster PV of the distributed PV itself.

[0042] Specifically, for a distributed photovoltaic, a spatial correlation calculation is performed based on the first historical photovoltaic information of the distributed photovoltaic and the second historical photovoltaic information of each centralized photovoltaic. Based on the calculated spatial correlation, the centralized photovoltaic with the largest spatial correlation is selected as the target cluster photovoltaic of the distributed photovoltaic itself.

[0043] Optionally, in another optional embodiment of the present invention, determining the target cluster photovoltaic corresponding to each of the distributed photovoltaics in the centralized photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information includes:

[0044] Determine a first autocorrelation coefficient of the distributed photovoltaic based on the first power sequence of the distributed photovoltaic;

[0045] Determine a second autocorrelation coefficient of each of the distributed photovoltaics based on the second power sequence of each of the concentrated photovoltaics;

[0046] Performing spatial correlation calculation based on the first power sequence and the first autocorrelation coefficient of the distributed photovoltaic and the second power sequence and the second autocorrelation coefficient of each of the centralized photovoltaics to determine the spatial correlation coefficient between the distributed photovoltaics and each centralized photovoltaic;

[0047] The centralized photovoltaic corresponding to the spatial correlation coefficient with the highest value in the distributed photovoltaic is determined as the target cluster photovoltaic.

[0048] Among them, the first autocorrelation coefficient can be the variance of the first power sequence of distributed photovoltaics, which is calculated by the first power sequence and the average value of the first power sequence. The specific calculation method is as follows: the average value of the first power sequence is calculated, and the average value of the square of the deviation between the first power sequence and the average value of the first power sequence is taken as the first autocorrelation coefficient of the first power sequence.

[0049] Among them, the second autocorrelation coefficient can be the variance of the second power sequence of the centralized photovoltaic system, which is calculated by the second power sequence and the average value of the second power sequence. The specific calculation method is as follows: the average value of the second power sequence is calculated, and the average value of the square of the deviation between the second power sequence and the average value of the second power sequence is taken as the second autocorrelation coefficient of the second power sequence.

[0050] Among them, the spatial correlation coefficient can be used to represent the spatial correlation between a distributed photovoltaic and a centralized photovoltaic. The calculation process of the spatial correlation coefficient is as follows: the correlation coefficient between distributed photovoltaic and centralized photovoltaic is calculated by the first power sequence of distributed photovoltaic and the second power sequence of centralized photovoltaic, and the average value of the product of the deviations of the two sequences is calculated as the correlation coefficient between centralized photovoltaic and distributed photovoltaic. After the correlation coefficient between centralized photovoltaic and distributed photovoltaic is calculated, the spatial correlation coefficient is calculated by combining the first autocorrelation coefficient and the second autocorrelation coefficient to determine the spatial correlation coefficient. Exemplarily, the formula of the spatial correlation coefficient is as follows:

[0051]

[0052] Among them, r xy is the spatial correlation coefficient, C xy (k) is the correlation coefficient between centralized photovoltaic and distributed photovoltaic, C xx (0) is the first autocorrelation coefficient, C yy(0) is the second autocorrelation coefficient, k and 0 represent the time series.

[0053] Specifically, the first autocorrelation coefficient of distributed photovoltaics is determined based on the first power sequence of distributed photovoltaics, the second autocorrelation coefficient of each distributed photovoltaic is determined based on the second power sequence of each centralized photovoltaic, spatial correlation calculation is performed according to the first power sequence and first autocorrelation coefficient of distributed photovoltaics and the second power sequence and second autocorrelation coefficient of each centralized photovoltaic, the spatial correlation coefficient of distributed photovoltaics and each centralized photovoltaic is determined, and for each distributed photovoltaic, the centralized photovoltaic with the highest spatial correlation coefficient between distributed photovoltaics and each centralized photovoltaic is selected as the target cluster photovoltaic.

[0054] S130 , predicting photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station using a pre-trained photovoltaic prediction model, and determining a day-ahead power curve.

[0055] The photovoltaic prediction model can be a neural network model obtained by training the long short-term memory network based on the photovoltaic power data and the meteorological data. It should be noted that the photovoltaic power data can be the power output of the photovoltaic system, and the meteorological data can be information such as light intensity and temperature.

[0056] The day-ahead power curve may be a day-ahead power curve constructed by outputting the day-ahead power generation per hour by a photovoltaic prediction model.

[0057] Optionally, the training process of the photovoltaic prediction model is as follows: perform maximum and minimum normalization on the photovoltaic power data and meteorological data to obtain a power training data set. Build a model based on the long short-term memory network, set the number of neural network layers, the number of long short-term memory neural network model units in each layer, the structure of the fully connected layer, the output layer, the mean square error function and the optimizer, and determine the power prediction model; divide the power training data set into a model training set, a model verification set and a model test set, train the power prediction model with the model training set, optimize the model parameters with the back propagation algorithm, evaluate the model performance with the model verification set, adjust the hyperparameters to improve the prediction accuracy, obtain the trained photovoltaic prediction model, and predict the power output of the photovoltaic prediction model with the model test set, generate the predicted photovoltaic power output based on the current meteorological data input, and use the mean square error to evaluate the performance of the photovoltaic power output. The formula for the mean square error performance evaluation is as follows:

[0058]

[0059] Among them, e RMSE is the mean square error of photovoltaic power output; x(i) is the measured power value of the model test set; is the photovoltaic power output, and N is the data of the model test set.

[0060] Specifically, for the virtual power station, the virtual power station power generation information of the virtual power station will be input into a pre-trained photovoltaic prediction model to perform photovoltaic power generation prediction, and obtain the day-ahead power curve output by the photovoltaic prediction model.

[0061] S140, collecting photovoltaic scene data of the virtual power station, and predicting according to the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine target photovoltaic scene data.

[0062] The photovoltaic scene data may be data information of the photovoltaic scene in which the centralized photovoltaic and distributed photovoltaic in the virtual power station are located. Exemplarily, the photovoltaic scene data includes the geographical location of the centralized photovoltaic and distributed photovoltaic in the virtual power station and the meteorological scene information corresponding to the geographical location, and the meteorological scene information includes light intensity and temperature.

[0063] Optionally, for each virtual power station, field data is collected on the installation locations of centralized photovoltaics and distributed photovoltaics in the virtual power station, and the geographical locations of the installation locations of centralized photovoltaics and distributed photovoltaics are spatially aggregated to obtain the geographical location of the virtual power station; the meteorological data of centralized photovoltaics and distributed photovoltaics are time series aggregated to obtain meteorological scene information.

[0064] Optionally, when predicting through the conditional generative adversarial network, a high-dimensional noise vector is generated through the standard normal distribution, and the photovoltaic scene data is used as the conditional information of the conditional generative adversarial network. The generator and discriminator of the conditional generative adversarial network are used to predict and generate the power curve and the high-dimensional noise vector, and the generator loss function is set to improve the generation ability, and the discriminator loss function is set to improve the identification ability to obtain the target photovoltaic scene data. Among them, the target photovoltaic scene data is the meteorological scene information of the distributed photovoltaic in the geographical location generated by the conditional generative network.

[0065] Specifically, for a virtual power station, photovoltaic scene data of the virtual power station is collected, and the photovoltaic scene data and the day-ahead power curve are predicted based on the photovoltaic scene data and the day-ahead power curve of the virtual power station through a conditional generative adversarial network to determine the target photovoltaic scene data.

[0066] The technical solution of the embodiment of the present invention, by acquiring the first historical photovoltaic information of distributed photovoltaic and the second historical photovoltaic information of centralized photovoltaic, the historical photovoltaic information is used as the basis for prediction, which can make the prediction result closer to reality and improve the rationality of the prediction result; according to the first historical photovoltaic information and the second historical photovoltaic information, the virtual power station and the virtual power station power generation information corresponding to the distributed photovoltaic are determined, and the utilization rate and flexibility of data calculation are improved through the data sharing of distributed photovoltaic and centralized photovoltaic, which can effectively reduce the calculation complexity and indirectly improve the efficiency of prediction; through the pre-trained photovoltaic prediction model, the photovoltaic power generation of the virtual power station is predicted according to the virtual power station power generation information, and the day-ahead power curve is determined, which can accurately capture each dependency in the historical information, and improve the prediction accuracy through the powerful feature extraction ability of the long-term and short-term neural network; the photovoltaic scene data of the virtual power station is collected, and the photovoltaic scene data and the day-ahead power curve of the virtual power station are predicted through the conditional generative adversarial network to determine the target photovoltaic scene data. The conditional generative adversarial network can effectively improve the quality of the day-ahead scene, and solve the technical problem that the distributed photovoltaic day-ahead scene cannot be accurately predicted in the prior art. It is possible to predict the day-ahead scenarios of each distributed photovoltaic system, improve the prediction quality of the day-ahead scenarios of distributed photovoltaic systems, and accurately predict the power in the distributed terminals corresponding to the distributed photovoltaic systems.

[0067] Figure 2 This is a flowchart of another distributed photovoltaic day-ahead scene prediction method provided by an embodiment of the present invention. The relationship between this embodiment and the above embodiment is a specific method for predicting photovoltaic scene data of a virtual power station through a conditional generative adversarial network. Figure 2 As shown, the method includes:

[0068] S210, obtaining first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics.

[0069] S220. Determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information.

[0070] S230 , predicting photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station using a pre-trained photovoltaic prediction model, and determining a day-ahead power curve.

[0071] S240, collecting photovoltaic scene data of the virtual power station, and predicting according to the photovoltaic scene data of the virtual power station and the day-ahead power curve through the conditional generative adversarial network to determine the day-ahead photovoltaic scene data.

[0072] Among them, the day-ahead photovoltaic scene data can be the meteorological scene information of the virtual power station at the geographical location generated by the conditional generation network.

[0073] Specifically, for a virtual power station, photovoltaic scene data of the virtual power station is collected, and the photovoltaic scene data and the day-ahead power curve are predicted based on the photovoltaic scene data and the day-ahead power curve of the virtual power station through a conditional generative adversarial network to determine the day-ahead photovoltaic scene data.

[0074] S250: Perform power decomposition on the day-ahead power curve to determine a first predicted power curve for the distributed photovoltaic system.

[0075] The first predicted power curve may be a power curve obtained by the photovoltaic prediction model for distributed photovoltaics. It should be noted that the photovoltaic prediction model does not directly predict the power generation of distributed photovoltaics, but is obtained by decomposing the day-ahead power curve, and the power curve of distributed photovoltaics is the power curve of all distributed photovoltaics in the virtual power station cluster.

[0076] Specifically, by performing power decomposition on the day-ahead power curve of the virtual power station, a first predicted power curve corresponding to the power generation power predicted by the photovoltaic prediction model of the distributed photovoltaic is obtained.

[0077] Optionally, in another optional embodiment of the present invention, performing power decomposition on the day-ahead power curve to determine the first predicted power curve of the distributed photovoltaic system includes:

[0078] The day-ahead power curve is decomposed based on the first power sequence and the second power sequence to obtain the power generation capacity ratio of the virtual power station; the day-ahead power curve is power-decomposed according to the power generation capacity ratio to determine the first predicted power curve.

[0079] Among them, the power generation capacity ratio can be the power generation ratio of a distributed photovoltaic cluster composed of multiple distributed photovoltaics in the virtual power station and a centralized photovoltaic. It should be noted that the power generation of a distributed photovoltaic can be obtained through the first power sequence and time of the distributed photovoltaic, the power generation of the distributed photovoltaic cluster is the sum of the power generation of all distributed photovoltaics, and the power generation of centralized photovoltaics can be obtained through the second power sequence and time of centralized photovoltaics. It should be noted that the time length selected when calculating the power generation of distributed photovoltaics and centralized photovoltaics is the same. Since it is a prediction of the day-ahead scene data, a period from the historical time to the day before the prediction is selected as the calculation time length for calculating the power generation. Exemplarily, the power generation capacity ratio can be expressed by k i-1 where i represents the time to be predicted for the virtual power station, and the power generation of the distributed photovoltaic cluster can be Q PV,i-1 , the power generation of concentrated photovoltaic can be Q DG,i-1 , the specific calculation formula is as follows:

[0080]

[0081] Optionally, after obtaining the power generation capacity ratio, the day-ahead power curve is decomposed based on the relationship between the power generation capacity ratio and the power sequence of the distributed photovoltaic cluster and the power sequence of the centralized photovoltaic cluster to obtain a first predicted power curve of the first distributed photovoltaic cluster. The power sequence of concentrated photovoltaic is expressed as The relationship between the power generation capacity ratio and the power sequence of the distributed photovoltaic cluster and the power sequence of the centralized photovoltaic is expressed as:

[0082]

[0083] S260: Perform autocorrelation calculation based on the first predicted power curve and the day-ahead photovoltaic scene data to determine target evaluation parameters.

[0084] Among them, the target evaluation parameter can be data used to evaluate the quality of the day-ahead photovoltaic scene data. It should be noted that since the day-ahead photovoltaic scene data cannot be specifically evaluated through parameters, power prediction is performed based on the day-ahead photovoltaic scene data to obtain the power generation power curve of the distributed photovoltaic cluster in the day-ahead photovoltaic scene data, and then autocorrelation calculation is performed with the first predicted power curve to obtain the correlation between the first predicted power curve and the day-ahead photovoltaic scene data as the target evaluation parameter.

[0085] Optionally, in another optional embodiment of the present invention, the autocorrelation calculation is performed based on the first predicted power curve and the day-ahead photovoltaic scene data to determine the target evaluation parameters, including: performing power prediction based on the first historical photovoltaic information and the day-ahead photovoltaic scene data to determine the second predicted power curve; and performing autocorrelation calculation based on the first predicted power curve and the second predicted power curve to determine the target evaluation parameters.

[0086] Optionally, when performing power prediction for the day-ahead photovoltaic scene data, a first meteorological data similar to the day-ahead photovoltaic scene data is selected from the first historical photovoltaic information of each distributed photovoltaic in the distributed photovoltaic cluster, and the power curve of each distributed photovoltaic corresponding to the first meteorological data is obtained from the first power curve of each distributed photovoltaic, and the power curve of each distributed photovoltaic is summed, and the calculation result is used as the second predicted power curve. The second predicted power curve can be a power curve for predicting photovoltaic power generation of the distributed photovoltaic under the day-ahead photovoltaic scene data.

[0087] Optionally, after obtaining the second predicted power curve corresponding to the day-ahead photovoltaic scene data in the distributed photovoltaic cluster, the second predicted power curve and the first predicted power curve are autocorrelated to obtain a target evaluation parameter. The specific calculation method is as follows: respectively calculate the means of the first predicted power curve and the second predicted power curve; respectively calculate the autocovariance of the first predicted power curve and the second predicted power curve; respectively calculate the variance of the first predicted power curve and the second predicted power curve, calculate the autocorrelation coefficient of the first predicted power curve based on the autocovariance and variance of the first predicted power curve, and calculate the autocorrelation coefficient of the second predicted power curve based on the autocovariance and variance of the second predicted power curve; calculate the difference in autocorrelation coefficients based on the autocorrelation coefficients of the first predicted power curve and the autocorrelation coefficients of the second predicted power curve, and use the difference in autocorrelation coefficients as the target evaluation parameter.

[0088] S270: If the target evaluation parameter meets a preset evaluation threshold, determining the day-ahead photovoltaic scene data as the target photovoltaic scene data.

[0089] The preset evaluation threshold may be a data threshold pre-set for judging the quality of the photovoltaic scene data on the day before. It should be noted that the smaller the difference in the autocorrelation coefficient, the higher the quality of the photovoltaic scene data on the day before, and the smaller the preset evaluation threshold, the better it can reflect the quality of the photovoltaic scene data on the day before. Exemplarily, the preset evaluation threshold is 0.02.

[0090] Specifically, the target evaluation parameter is numerically compared with the preset evaluation threshold. If the target evaluation parameter is greater than the preset evaluation threshold, it is considered that the target evaluation parameter does not meet the preset evaluation threshold, indicating that the quality of the day-ahead photovoltaic scene data is too low, and a conditional generative adversarial network is required to continue generating day-ahead photovoltaic scene data; if the target evaluation parameter is less than the preset evaluation threshold, it is considered that the target evaluation parameter meets the preset evaluation threshold, and the day-ahead photovoltaic scene data is determined as the target photovoltaic scene data of each distributed photovoltaic in the distributed photovoltaic cluster corresponding to the virtual power station.

[0091] The technical solution of the embodiment of the present invention makes the target photovoltaic scene data more consistent with the real scene of distributed photovoltaics by judging the quality of the day-ahead photovoltaic scene data, improves the prediction quality of the day-ahead scene of distributed photovoltaics, and effectively performs self-power prediction in the distributed terminals corresponding to distributed photovoltaics.

[0092] Figure 3 The following is a schematic diagram of the structure of a distributed photovoltaic day-ahead scene prediction device provided by an embodiment of the present invention. Figure 3 As shown, the device includes: a data acquisition module 310, a virtual power station combination module 320, a power curve prediction module 330 and a photovoltaic day-ahead scene prediction module 340; wherein,

[0093] The data acquisition module 310 is used to acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics;

[0094] A virtual power station combination module 320, configured to determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information;

[0095] A power curve prediction module 330 is used to predict photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station through a pre-trained photovoltaic prediction model, and determine a day-ahead power curve;

[0096] The photovoltaic day-ahead scene prediction module 340 is used to collect photovoltaic scene data of the virtual power station, and to predict the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine the target photovoltaic scene data.

[0097] The technical solution of the embodiment of the present invention, by acquiring the first historical photovoltaic information of distributed photovoltaic and the second historical photovoltaic information of centralized photovoltaic, the historical photovoltaic information is used as the basis for prediction, which can make the prediction result closer to reality and improve the rationality of the prediction result; according to the first historical photovoltaic information and the second historical photovoltaic information, the virtual power station and the virtual power station power generation information corresponding to the distributed photovoltaic are determined, and the utilization rate and flexibility of data calculation are improved through the data sharing of distributed photovoltaic and centralized photovoltaic, which can effectively reduce the calculation complexity and indirectly improve the efficiency of prediction; through the pre-trained photovoltaic prediction model, the photovoltaic power generation of the virtual power station is predicted according to the virtual power station power generation information, and the day-ahead power curve is determined, which can accurately capture each dependency in the historical information, and improve the prediction accuracy through the powerful feature extraction ability of the long-term and short-term neural network; the photovoltaic scene data of the virtual power station is collected, and the photovoltaic scene data and the day-ahead power curve of the virtual power station are predicted through the conditional generative adversarial network to determine the target photovoltaic scene data. The conditional generative adversarial network can effectively improve the quality of the day-ahead scene, and solve the technical problem that the distributed photovoltaic day-ahead scene cannot be accurately predicted in the prior art. It is possible to predict the day-ahead scenarios of each distributed photovoltaic system, improve the prediction quality of the day-ahead scenarios of distributed photovoltaic systems, and accurately predict the power in the distributed terminals corresponding to the distributed photovoltaic systems.

[0098] Optionally, the data acquisition module is specifically used for: the first historical photovoltaic information includes a first power sequence and first meteorological data; the second historical photovoltaic information includes a second power sequence and second meteorological data.

[0099] Optionally, the photovoltaic day-ahead scene prediction module is specifically used for:

[0100] Predicting the photovoltaic scene data of the virtual power station and the day-ahead power curve by the conditional generative adversarial network to determine the day-ahead photovoltaic scene data;

[0101] Performing power decomposition on the day-ahead power curve to determine a first predicted power curve of the distributed photovoltaic system;

[0102] Performing autocorrelation calculation based on the first predicted power curve and the day-ahead photovoltaic scene data to determine a target evaluation parameter;

[0103] If the target evaluation parameter meets a preset evaluation threshold, the day-ahead photovoltaic scene data is determined as the target photovoltaic scene data.

[0104] Optionally, the photovoltaic day-ahead scene prediction module is further used for:

[0105] Decomposing the day-ahead power curve based on the first power sequence and the second power sequence to obtain a power generation capacity ratio of the virtual power plant;

[0106] The day-ahead power curve is decomposed according to the power generation capacity ratio to determine the first predicted power curve.

[0107] Optionally, the photovoltaic day-ahead scene prediction module is further used for:

[0108] Perform power prediction based on the first historical photovoltaic information and the day-ahead photovoltaic scene data to determine a second predicted power curve;

[0109] An autocorrelation calculation is performed according to the first predicted power curve and the second predicted power curve to determine the target evaluation parameter.

[0110] Optionally, the virtual power plant combination module is specifically used for:

[0111] Determine a target cluster photovoltaic corresponding to each of the distributed photovoltaics in the centralized photovoltaics according to the first historical photovoltaic information and the second historical photovoltaic information;

[0112] The distributed photovoltaic cluster is clustered into the target photovoltaic cluster, and the virtual power station and the power generation information of the virtual power station are determined.

[0113] Optionally, the virtual power plant combination module is further used for:

[0114] Determine a first autocorrelation coefficient of the distributed photovoltaic based on the first power sequence of the distributed photovoltaic;

[0115] Determine a second autocorrelation coefficient of each of the distributed photovoltaics based on the second power sequence of each of the concentrated photovoltaics;

[0116] Performing spatial correlation calculation based on the first power sequence and the first autocorrelation coefficient of the distributed photovoltaic and the second power sequence and the second autocorrelation coefficient of each of the centralized photovoltaics to determine the spatial correlation coefficient between the distributed photovoltaics and each centralized photovoltaic;

[0117] The centralized photovoltaic corresponding to the spatial correlation coefficient with the highest value in the distributed photovoltaic is determined as the target cluster photovoltaic.

[0118] The distributed photovoltaic day-ahead scene prediction device provided in the embodiment of the present invention can execute the distributed photovoltaic day-ahead scene prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0119] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their modes are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0120] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0121] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer grid such as the Internet and / or various telecommunication grids.

[0122] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the day-ahead scenario prediction method for distributed photovoltaics.

[0123] In some embodiments, the day-ahead scene prediction method for distributed photovoltaics may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the day-ahead scene prediction method for distributed photovoltaics described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the day-ahead scene prediction method for distributed photovoltaics in any other appropriate manner (e.g., by means of firmware).

[0124] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0125] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the modes / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0126] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0127] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0128] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a grid browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication grid). Examples of communication grids include: a local area network (LAN), a wide area network (WAN), a blockchain grid, and the Internet.

[0129] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication grid. The client and server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0130] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0131] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the distributed photovoltaic day-ahead scene prediction method provided in any embodiment of the present invention are implemented. The method includes:

[0132] Acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics;

[0133] Determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information;

[0134] Predicting photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station by using a pre-trained photovoltaic prediction model to determine a day-ahead power curve;

[0135] Photovoltaic scene data of a virtual power station is collected, and prediction is performed based on the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine target photovoltaic scene data.

[0136] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0137] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0138] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0139] Computer program code for performing the operation of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of grid, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0140] It should be understood by those skilled in the art that the above modules or steps of the present invention can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a grid composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0141] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0142] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for day-ahead scene prediction of distributed photovoltaics, characterized in that: include: Acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics; Determine a virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information; Predicting photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station by using a pre-trained photovoltaic prediction model to determine a day-ahead power curve; Photovoltaic scene data of a virtual power station is collected, and prediction is performed based on the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine target photovoltaic scene data.

2. The method according to claim 1, characterized in that The first historical photovoltaic information includes a first power sequence and first meteorological data; the second historical photovoltaic information includes a second power sequence and second meteorological data.

3. The method according to claim 2, characterized in that The predicting by the conditional generative adversarial network according to the photovoltaic scene data of the virtual power station and the day-ahead power curve to determine the target photovoltaic scene data includes: Predicting the photovoltaic scene data of the virtual power station and the day-ahead power curve by the conditional generative adversarial network to determine the day-ahead photovoltaic scene data; Performing power decomposition on the day-ahead power curve to determine a first predicted power curve of the distributed photovoltaic system; Performing autocorrelation calculation based on the first predicted power curve and the day-ahead photovoltaic scene data to determine a target evaluation parameter; If the target evaluation parameter meets a preset evaluation threshold, the day-ahead photovoltaic scene data is determined as the target photovoltaic scene data.

4. The method according to claim 3, characterized in that The performing power decomposition on the day-ahead power curve to determine the first predicted power curve of the distributed photovoltaic power generation includes: Decomposing the day-ahead power curve based on the first power sequence and the second power sequence to obtain a power generation capacity ratio of the virtual power plant; The day-ahead power curve is decomposed according to the power generation capacity ratio to determine the first predicted power curve.

5. The method according to claim 4, characterized in that The step of performing autocorrelation calculation based on the first predicted power curve and the day-ahead photovoltaic scene data to determine a target evaluation parameter includes: Perform power prediction based on the first historical photovoltaic information and the day-ahead photovoltaic scene data to determine a second predicted power curve; An autocorrelation calculation is performed according to the first predicted power curve and the second predicted power curve to determine the target evaluation parameter.

6. The method according to claim 2, characterized in that The determining the virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information includes: Determine a target cluster photovoltaic corresponding to each of the distributed photovoltaics in the centralized photovoltaics according to the first historical photovoltaic information and the second historical photovoltaic information; The distributed photovoltaic cluster is clustered into the target photovoltaic cluster, and the virtual power station and the power generation information of the virtual power station are determined.

7. The method according to claim 6, characterized in that The step of determining the target cluster photovoltaic corresponding to each of the distributed photovoltaics in the centralized photovoltaic system according to the first historical photovoltaic information and the second historical photovoltaic information includes: Determine a first autocorrelation coefficient of the distributed photovoltaic based on the first power sequence of the distributed photovoltaic; Determine a second autocorrelation coefficient of each of the distributed photovoltaics based on the second power sequence of each of the concentrated photovoltaics; Performing spatial correlation calculation based on the first power sequence and the first autocorrelation coefficient of the distributed photovoltaic and the second power sequence and the second autocorrelation coefficient of each of the centralized photovoltaic, determining the spatial correlation coefficient between the distributed photovoltaic and each centralized photovoltaic; The centralized photovoltaic corresponding to the spatial correlation coefficient with the highest value in the distributed photovoltaic is determined as the target cluster photovoltaic.

8. A distributed photovoltaic day-ahead scene prediction device, characterized in that: include A data acquisition module, used to acquire first historical photovoltaic information of distributed photovoltaics and second historical photovoltaic information of centralized photovoltaics; A virtual power station combination module, used to determine the virtual power station and virtual power station power generation information corresponding to the distributed photovoltaic according to the first historical photovoltaic information and the second historical photovoltaic information; A power curve prediction module, used to predict photovoltaic power generation of the virtual power station according to the power generation information of the virtual power station through a pre-trained photovoltaic prediction model, and determine a day-ahead power curve; The photovoltaic day-ahead scene prediction module is used to collect photovoltaic scene data of the virtual power station, and predict the photovoltaic scene data of the virtual power station and the day-ahead power curve through a conditional generative adversarial network to determine the target photovoltaic scene data.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the day-ahead scene prediction method for distributed photovoltaics according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the day-ahead scene prediction method for distributed photovoltaics according to any one of claims 1 to 7 when executed.

Images