Prediction method of regional surface radiation intensity based on cloud movement trajectory

Abstract

The embodiment of the present application provides a method for predicting regional surface radiation intensity based on cloud movement trajectories. The method comprises: obtaining the current satellite cloud image of the photovoltaic power station to be predicted and the weather forecast data of the area to be predicted where the photovoltaic power station to be predicted is located, determining the cloud movement pattern of the area to be predicted based on the satellite cloud image and the weather forecast data, determining the predicted displacement vector of the cloud in the area to be predicted based on the cloud displacement vector calculation model, determining the predicted cloud map of the area to be predicted based on the predicted displacement vector, the predicted cloud map including the predicted position of the cloud in the area to be predicted, determining the target area with influence based on the geographical location of the photovoltaic power station to be predicted, determining the solar irradiance depreciation coefficient of the target area, and determining the surface radiation intensity of the area to be predicted at a preset time in the future based on the historical surface measured radiation intensity of the area to be predicted under historical clear sky conditions at a preset time in the future and the solar irradiance depreciation coefficient.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic prediction, and specifically to a method, storage medium, and processor for predicting regional surface radiation intensity based on cloud movement trajectories. Background Art

[0002] With the rapid development of society and economy worldwide, humanity's demand for energy continues to increase. The traditional fossil fuel structure, dominated by oil and coal, is causing environmental damage such as global warming. The development and utilization of clean energy sources such as solar energy has attracted widespread attention. Photovoltaic power generation is the primary means of converting solar energy into electricity, but its output power is affected by factors such as surface irradiance attenuation.

[0003] Existing research on surface irradiance attenuation prediction primarily relies on predicting cloud trajectories based on ground-based or satellite cloud images, and then predicting surface irradiance attenuation based on this prediction. This approach assumes that cloud shapes remain constant between adjacent cloud images, and then uses linear extrapolation to calculate cloud displacement vectors and velocities. However, existing techniques ignore the deformation and formation and dissipation of clouds, resulting in large prediction errors and high uncertainty. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a method, storage medium, and processor for predicting regional surface radiation intensity based on cloud movement trajectories.

[0005] To achieve the above objectives, the present application provides, in a first aspect, a method for predicting regional surface radiation intensity based on cloud movement trajectories, comprising:

[0006] Obtain the current satellite cloud image of the photovoltaic power station to be predicted and the weather forecast data of the area to be predicted where the photovoltaic power station to be predicted is located;

[0007] Determine the cloud movement pattern in the area to be predicted based on current satellite cloud imagery and weather forecast data;

[0008] Using the cloud displacement vector calculation model, the predicted displacement vector of the cloud in the predicted area is determined according to the cloud movement pattern in the predicted area;

[0009] Determining a predicted cloud map of the area to be predicted based on the current position of the cloud cluster in the area to be predicted and the predicted displacement vector, wherein the predicted cloud map includes the predicted position of the cloud cluster in the area to be predicted;

[0010] Determine the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted based on the geographical location of the photovoltaic power station to be predicted;

[0011] Determine the solar irradiance loss factor of the target area;

[0012] The surface radiation intensity of the area to be predicted at a preset time in the future is determined based on the historical surface measured radiation intensity under historical clear sky conditions in the area to be predicted at a preset time in the future and the solar irradiance reduction coefficient.

[0013] In an embodiment of the present application, determining the solar irradiance loss coefficient of the target area includes: extracting the grayscale value of each pixel in the target area; determining the grayscale mean of the target area based on the grayscale value of each pixel; and determining the solar irradiance loss coefficient corresponding to the cloud map in the historical data that has the same grayscale mean and weather data close to the weather forecast data of the area to be predicted as the solar irradiance loss coefficient corresponding to the target area.

[0014] In the embodiment of the present application, the solar irradiance loss coefficient λ corresponding to the cloud map in the historical data is calculated according to the following formula (1):

[0015]

[0016] Among them, I0 is the measured surface radiation intensity, and I1 is the measured surface radiation intensity under clear sky conditions.

[0017] In an embodiment of the present application, determining the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted based on the geographical location of the photovoltaic power station to be predicted includes: determining the solar azimuth and solar altitude angle based on the geographical location of the photovoltaic power station to be predicted; determining the intersection of the sunlight and the cloud cluster in the prediction cloud map based on the solar azimuth and solar altitude angle, and selecting an area around the intersection that is consistent with the area of the photovoltaic power station to be predicted as the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted, with the intersection as the center.

[0018] In an embodiment of the present application, determining the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data includes: obtaining historical satellite cloud image sequence images and historical weather data of the photovoltaic power station; after preprocessing the historical satellite cloud image sequence images, dividing the preprocessed historical satellite cloud image sequence images into multiple image pairs, wherein each image pair includes two sequence images of adjacent time, for each image pair, extracting the feature vector of the image pair and the weather data corresponding to the image pair, using a clustering algorithm, classifying the multiple image pairs according to the weather data corresponding to each image pair to determine the cloud movement patterns corresponding to the multiple image pairs, establishing a cloud movement pattern recognition model according to the cloud movement patterns corresponding to the multiple image pairs, and using the cloud movement pattern recognition model to determine the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data.

[0019] In an embodiment of the present application, before using a cloud displacement vector calculation model to determine the predicted displacement vector of the cloud in the to-be-predicted area according to the cloud motion pattern of the to-be-predicted area, for each image pair, the cloud displacement vector of the image pair is determined separately based on a plurality of cloud trajectory tracking algorithms; for each image pair, the weight combination of the image pair is determined according to the cloud displacement vectors determined by all the cloud trajectory tracking algorithms; for each cloud motion pattern, the optimal weight combination of the cloud motion pattern is determined according to the weight combination of each image pair under the cloud motion pattern, so as to establish a cloud displacement vector calculation model.

[0020] In an embodiment of the present application, for each image pair, determining the cloud displacement vector of each image pair based on multiple cloud trajectory tracking algorithms includes calculating the cloud displacement vector of each image pair according to formula (2):

[0021]

[0022] X and Y represent the displacement vector of the cloud in the image pair in the X direction and the Y direction, respectively. f1, f2, ..., f n Represents different weights, V1, V2, ..., V n represents different cloud trajectory tracking algorithms, V i (x) represents the displacement vector of the cloud in the X direction calculated by the ith cloud trajectory tracking algorithm, V i (y) represents the displacement vector of the cloud in the Y direction calculated using the i-th cloud trajectory tracking algorithm.

[0023] In an embodiment of the present application, for each cloud motion mode, the optimal weight combination of the cloud motion mode is determined according to the weight combination of each image pair under the cloud motion mode to establish a cloud displacement vector calculation model, including: determining all image pairs under each cloud motion mode, and the weight combination of each image pair; for each cloud motion mode, determining the mean of each weight in the weight combination according to the weight combination of all image pairs under the cloud motion mode; for each cloud motion mode, determining the weight combination composed of the mean of each weight as the optimal weight combination of the cloud motion mode.

[0024] A second aspect of the present application provides a processor configured to execute the above-mentioned method for predicting regional surface radiation intensity based on cloud movement trajectories.

[0025] A third aspect of the present application provides a machine-readable storage medium having instructions stored thereon, which, when executed by a processor, configures the processor to execute the above-mentioned method for predicting regional surface radiation intensity based on cloud movement trajectories.

[0026] The above-mentioned method for predicting regional surface radiation intensity based on cloud motion trajectories obtains current satellite cloud imagery of the photovoltaic power station to be predicted and weather forecast data for the region to be predicted where the photovoltaic power station is located. The cloud motion pattern of the region to be predicted is determined based on the current satellite cloud imagery and weather forecast data. A cloud displacement vector calculation model is used to determine the predicted displacement vector of the clouds in the region to be predicted based on the cloud motion pattern. A predicted cloud map for the region to be predicted is determined based on the current position and predicted displacement vector of the clouds in the region to be predicted. The predicted cloud map includes the predicted position of the clouds in the region to be predicted. Target areas in the predicted cloud map that may affect the photovoltaic power station to be predicted are determined based on the geographic location of the photovoltaic power station to be predicted. The solar irradiance loss coefficient for the target areas is determined. The surface radiation intensity of the region to be predicted at a preset future time is determined based on the historically measured surface radiation intensity under historical clear sky conditions in the region to be predicted and the solar irradiance loss coefficient. This method can effectively predict complex motion processes such as cloud deformation and formation and dissipation, more accurately predict cloud motion trajectories, improve the accuracy of surface irradiance attenuation prediction, and ensure the safe and stable operation of the power grid system.

[0027] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present application but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:

[0029] Figure 1 The following schematically illustrates a flow chart of a method for predicting regional surface radiation intensity based on cloud movement trajectories according to an embodiment of the present application;

[0030] Figure 2 The following schematically shows a predicted cloud movement trajectory according to an embodiment of the present application;

[0031] Figure 3 The internal structure diagram of a computer device according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0032] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0033] Figure 1 The following schematically shows a flow chart of a method for predicting regional surface radiation intensity based on cloud movement trajectories according to an embodiment of the present application. Figure 1 As shown, in one embodiment of the present application, a method for predicting regional surface radiation intensity based on cloud movement trajectories is provided, comprising the following steps:

[0034] Step 101: Obtain the current satellite cloud image of the photovoltaic power station to be predicted and the weather forecast data of the area to be predicted where the photovoltaic power station to be predicted is located.

[0035] Step 102: Determine the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data.

[0036] Step 103 : using a cloud displacement vector calculation model, the predicted displacement vector of the cloud in the area to be predicted is determined according to the cloud motion pattern in the area to be predicted.

[0037] Step 104 : determining a predicted cloud map of the area to be predicted based on the current position of the cloud cluster in the area to be predicted and the predicted displacement vector, wherein the predicted cloud map includes the predicted position of the cloud cluster in the area to be predicted.

[0038] Step 105 : determining a target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted according to the geographical location of the photovoltaic power station to be predicted.

[0039] Step 106: Determine the solar irradiance loss coefficient of the target area.

[0040] Step 107 , determining the surface radiation intensity of the area to be predicted at a preset time in the future based on the historical surface measured radiation intensity and the solar irradiance depreciation coefficient of the area to be predicted under historical clear sky conditions at a preset time in the future.

[0041] Satellite cloud images are cloud images captured and transmitted back to Earth by meteorological satellites. They can display cloud distribution and structure over a wide area. Weather forecast data includes cloud cover at different cloud heights, wind speed, wind direction, and relative humidity. The current satellite cloud image of the photovoltaic power station to be predicted, as well as weather forecast data for the area where the photovoltaic power station is located, is obtained. Preprocessing involves mosaicking, cropping, and histogram equalization. This preprocessing generates a sequence of current satellite cloud images of the photovoltaic power station to be predicted. Two images from adjacent time periods in the sequence are grouped together to generate multiple image pairs. For each image pair, a feature vector is extracted, along with the corresponding weather data. The feature vectors of an image pair include texture features, colorimetric features, information features, and frequency domain features. Texture features refer to surface characteristics of the image pair; colorimetric features refer to the distribution and depth of various colors in the image pair; information features output information about the image pair; and frequency domain features refer to the characteristics of the frequency fluctuations in the image pair. Historical satellite cloud image sequences and historical weather data for the photovoltaic power station to be predicted are obtained, preprocessed, and divided into multiple image pairs, each consisting of two sequential images from adjacent time periods. For each image pair, the feature vector and corresponding weather data are extracted. A clustering algorithm is used to classify the multiple image pairs based on the historical weather data corresponding to each image pair obtained from the preprocessed historical satellite cloud image sequences to determine the cloud motion patterns corresponding to the multiple image pairs. A cloud motion pattern recognition model is established based on the cloud motion patterns corresponding to the multiple image pairs. The cloud motion pattern recognition model is then used to determine the cloud motion pattern for the predicted area based on the current satellite cloud imagery and weather forecast data for the predicted photovoltaic power station.

[0042] For multiple image pairs divided from preprocessed historical satellite cloud image sequences, cloud displacement vectors were calculated for each pair using various traditional cloud trajectory tracking algorithms. A linear regression model was then fitted to obtain the optimal weight combination for each image pair. The optimal weights for each motion mode were averaged to obtain the optimal weight corresponding to each motion mode. A cloud displacement vector calculation model was established. For each image pair, the cloud displacement vector was determined using various cloud trajectory tracking algorithms.

[0043] The current satellite cloud imagery of the photovoltaic power station to be predicted is divided into multiple image pairs. Cloud displacement vectors are calculated for each of these image pairs using various traditional cloud trajectory tracking algorithms. A linear regression model is then used to fit the optimal weighted combination for each image pair. For each cloud motion pattern, the mean of each weight in the weighted combination is determined based on the weighted combination of all image pairs under that cloud motion pattern. For each cloud motion pattern, the weighted combination consisting of the mean of each weight is determined as the optimal weighted combination for that cloud motion pattern, thereby establishing a cloud displacement vector calculation model. Using the cloud displacement vector calculation model, the predicted displacement vectors of the clouds in the predicted area are determined based on the cloud motion pattern. Based on the current cloud positions and predicted displacement vectors, a predicted cloud map is generated for the predicted area. The predicted cloud map includes the predicted locations of the clouds in the predicted area, and the geographic location information of the photovoltaic power station to be predicted is obtained. This geographic location information includes longitude, latitude, and altitude. The solar azimuth and altitude angles are determined based on the geographic location of the PV power station to be predicted. The intersection of the sun's rays and the cloud clusters in the predicted cloud map is determined based on the sun's azimuth and altitude angles for the PV power station to be predicted. A region around the intersection, centered on the intersection of the sun's rays and the cloud clusters, corresponding to the area of the PV power station to be predicted, is selected as the target region in the predicted cloud map that will affect the PV power station to be predicted. The grayscale value of each pixel in the target region is extracted. Grayscale refers to the color depth of a pixel in a black-and-white image, with white being 255 and black being 0. The grayscale values of the target region are compared with the historical data of the PV power station to be predicted. The solar irradiance loss coefficient corresponding to the cloud map with the same grayscale mean value in the historical data and weather data close to the weather forecast data for the target region is determined as the solar irradiance loss coefficient for the target region. The surface radiation intensity of the target region at a preset future time is determined based on the historically measured surface radiation intensity under historical clear sky conditions in the target region and the solar irradiance loss coefficient. Through this method, we can effectively predict complex movement processes such as cloud deformation, formation and disappearance, more accurately predict cloud movement trajectories, improve the accuracy of surface irradiance attenuation prediction, and ensure the safe and stable operation of the power grid system.

[0044] In one embodiment, determining the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data includes: obtaining historical satellite cloud image sequence images and historical weather data of the photovoltaic power station; after preprocessing the historical satellite cloud image sequence images, dividing the preprocessed historical satellite cloud image sequence images into multiple image pairs, wherein each image pair includes two sequence images at adjacent times; for each image pair, extracting the feature vector of the image pair and the weather data corresponding to the image pair; using a clustering algorithm, classifying the multiple image pairs according to the weather data corresponding to each image pair to determine the cloud movement patterns corresponding to the multiple image pairs; establishing a cloud movement pattern recognition model according to the cloud movement patterns corresponding to the multiple image pairs; and using the cloud movement pattern recognition model, determining the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data.

[0045] By preprocessing the current satellite cloud imagery of the PV power plant to be predicted, a sequence of satellite cloud images of the plant is generated. Two images from adjacent time periods in the sequence are grouped together to produce multiple image pairs. For each image pair, a feature vector is extracted, along with the weather data corresponding to the current image pair. The feature vectors of an image pair include texture features, chromaticity features, information theory features, and frequency domain features. Texture features refer to the surface characteristics of the image pair, chromaticity features refer to the distribution and depth of the various colors in the image pair, information theory features output information about the image pair, and frequency domain features refer to the characteristics of the frequency fluctuations of the image pair.

[0046] Historical satellite cloud image sequences and historical weather data for the photovoltaic power station to be predicted are obtained, preprocessed, and divided into multiple image pairs, each consisting of two sequential images from adjacent time periods. For each image pair, the feature vector and corresponding weather data are extracted. A clustering algorithm is used to classify the multiple image pairs based on the historical weather data corresponding to each image pair obtained from the preprocessed historical satellite cloud image sequences to determine the cloud motion patterns corresponding to the multiple image pairs. A cloud motion pattern recognition model is established based on the cloud motion patterns corresponding to the multiple image pairs. The cloud motion pattern recognition model is then used to determine the cloud motion pattern for the predicted area based on the current satellite cloud imagery and weather forecast data for the predicted photovoltaic power station.

[0047] In one embodiment, before using a cloud displacement vector calculation model to determine the predicted displacement vector of the cloud in the to-be-predicted area according to the cloud motion pattern of the to-be-predicted area, for each image pair, the cloud displacement vector of the image pair is determined separately based on multiple cloud trajectory tracking algorithms. For each image pair, the weight combination of the image pair is determined according to the cloud displacement vectors determined by all the cloud trajectory tracking algorithms. For each cloud motion pattern, the optimal weight combination of the cloud motion pattern is determined according to the weight combination of each image pair under the cloud motion pattern to establish a cloud displacement vector calculation model.

[0048] For multiple image pairs divided from the preprocessed historical satellite cloud image sequence, the cloud displacement vectors are calculated for each image pair based on a variety of traditional cloud trajectory tracking algorithms. The optimal weight combination of each image pair is obtained by fitting the linear regression model. The optimal weights of the image pairs contained in each motion mode are averaged to obtain the optimal weight corresponding to each motion mode. A cloud displacement vector calculation model has been established.

[0049] In one embodiment, for each image pair, determining the cloud displacement vector of each image pair based on multiple cloud trajectory tracking algorithms includes calculating the cloud displacement vector of each image pair according to formula (2):

[0050]

[0051] X and Y represent the displacement vector of the cloud in the image pair in the X direction and the Y direction, respectively. f1, f2, ..., f n Represents different weights, V1, V2, ..., V n represents different cloud trajectory tracking algorithms, V i (x) represents the displacement vector of the cloud in the X direction calculated by the ith cloud trajectory tracking algorithm, V i (y) represents the displacement vector of the cloud in the Y direction calculated using the i-th cloud trajectory tracking algorithm.

[0052] For the multiple image pairs divided by the pre-processed historical satellite cloud image sequence, for each image pair, the cloud displacement vector of the image pair is determined based on multiple cloud trajectory tracking algorithms, including calculating the cloud displacement vector of each image pair according to formula (2):

[0053]

[0054] X and Y represent the displacement vector of the cloud in the image pair in the X direction and the Y direction, respectively. f1, f2, ..., f n Represents different weights, V1, V2, ..., V n represents different cloud trajectory tracking algorithms, Vi (x) represents the displacement vector of the cloud in the X direction calculated by the ith cloud trajectory tracking algorithm, V i (y) represents the displacement vector of the cloud in the Y direction calculated using the i-th cloud trajectory tracking algorithm.

[0055] In one embodiment, for each cloud motion mode, the optimal weight combination of the cloud motion mode is determined according to the weight combination of each image pair under the cloud motion mode to establish a cloud displacement vector calculation model, including: determining all image pairs under each cloud motion mode, and the weight combination of each image pair; for each cloud motion mode, determining the mean of each weight in the weight combination according to the weight combination of all image pairs under the cloud motion mode; for each cloud motion mode, determining the weight combination composed of the mean of each weight as the optimal weight combination of the cloud motion mode.

[0056] The current satellite cloud image of the photovoltaic power station to be predicted is divided into multiple image pairs. The cloud displacement vectors of the multiple image pairs are calculated based on a variety of traditional cloud trajectory tracking algorithms. The optimal weight combination of each image pair is obtained by linear regression model fitting. For each cloud motion mode, the mean of each weight in the weight combination is determined according to the weight combination of all image pairs under the cloud motion mode; for each cloud motion mode, the weight combination composed of the mean of each weight is determined as the optimal weight combination of the cloud motion mode to establish a cloud displacement vector calculation model.

[0057] In one embodiment, determining the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted based on the geographical location of the photovoltaic power station to be predicted includes: determining the solar azimuth and solar altitude angle based on the geographical location of the photovoltaic power station to be predicted, determining the intersection of the sunlight and the cloud cluster in the prediction cloud map based on the solar azimuth and solar altitude angle, and selecting an area around the intersection that is consistent with the area of the photovoltaic power station to be predicted as the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted, with the intersection as the center.

[0058] Using the cloud displacement vector calculation model, the predicted displacement vector of the cloud in the predicted area is determined based on the cloud movement pattern in the predicted area. Based on the current position of the cloud in the predicted area and the predicted displacement vector, a predicted cloud map of the predicted area is determined. The predicted cloud map includes the predicted position of the cloud in the predicted area, and the geographical location information of the photovoltaic power station to be predicted is obtained. The geographical location information of the photovoltaic power station includes longitude and latitude information, altitude, etc. The solar azimuth and solar altitude angle are determined based on the geographical location of the photovoltaic power station to be predicted. The calculation formula for the solar altitude angle is: in is the latitude of the photovoltaic power station to be predicted, δ is the declination angle of the photovoltaic power station to be predicted, and ω is the hour angle of the photovoltaic power station to be predicted. The hour angle calculation formula is: ω=(t-12)×15 ° , the declination angle calculation formula: k is the serial number of a day in a year. The formula for calculating the solar azimuth angle is: δ is the declination angle of the photovoltaic power station to be predicted, is the latitude of the PV power station to be predicted, and α is the solar altitude angle of the PV power station to be predicted. The intersection point of the sunlight and the cloud cluster in the predicted cloud map is determined based on the solar azimuth and solar altitude of the PV power station to be predicted. The calculation formula for the intersection point of sunlight and cloud cluster is: Where H is the cloud height, α is the solar altitude angle at the PV plant to be predicted, and γ is the solar azimuth angle at the PV plant to be predicted. The intersection of the sunlight and the cloud cluster at the predicted PV plant is used as the center. An area around the intersection that matches the area of the PV plant to be predicted is selected as the target area in the prediction cloud map that may affect the PV plant to be predicted.

[0059] In one embodiment, determining the solar irradiance loss coefficient of the target area includes: extracting the grayscale value of each pixel in the target area; determining the grayscale mean of the target area based on the grayscale value of each pixel; and determining the solar irradiance loss coefficient corresponding to the cloud map in the historical data that has the same grayscale mean and weather data close to the weather forecast data of the area to be predicted as the solar irradiance loss coefficient corresponding to the target area.

[0060] Extract the grayscale value of each pixel in the target area. Grayscale refers to the color depth of a pixel in a black-and-white image, with white being 255 and black being 0. Compare the historical data of the PV power station to be predicted with the grayscale values of the target area. Determine the solar irradiance loss coefficient corresponding to the cloud map in the historical data that has the same grayscale mean value and weather data close to the weather forecast data for the target area as the solar irradiance loss coefficient for the target area.

[0061] In one embodiment, the solar irradiance loss coefficient λ corresponding to the cloud image in the historical data is calculated according to the following formula (1):

[0062]

[0063] Among them, I0 is the measured surface radiation intensity, and I1 is the measured surface radiation intensity under clear sky conditions.

[0064] The calculation formula for the historical impairment coefficient λ is: Where I0 is the measured surface radiation intensity, and I1 is the measured surface radiation intensity under clear sky conditions. The mathematical expectation of the historical radiation loss coefficient is used to obtain the solar irradiance loss coefficient corresponding to the target area. On clear days, λ = 1. The surface radiation intensity of the target area at a preset future time is determined based on the historical measured surface radiation intensity under clear sky conditions and the solar irradiance loss coefficient.

[0065] Figure 2 This is a schematic diagram of a method for predicting regional surface radiation intensity based on cloud motion trajectories. As shown in the figure, satellite cloud images and numerical weather forecasts are obtained for the photovoltaic power station to be predicted at times t1 and t2. The weather forecast data includes cloud cover at different cloud layer heights, wind speed, wind direction, and relative humidity. Feature vectors of satellite cloud image pairs at times t1 and t2, along with the corresponding weather data, are extracted. A clustering algorithm is then used to classify multiple image pairs based on the weather data corresponding to each satellite cloud image pair at times t1 and t2 to determine the cloud motion patterns corresponding to each image pair. A cloud motion pattern recognition model is then established based on the cloud motion patterns corresponding to these image pairs. The cloud motion pattern recognition model is then used to determine the cloud motion pattern for the predicted area based on the current satellite cloud imagery and weather forecast data. For each cloud motion pattern, the optimal weighted combination of each image pair within the cloud motion pattern is determined. The solar azimuth and altitude are determined based on the geographic location of the PV power station to be predicted. The intersection of the sun's rays and the clouds in the predicted cloud map is determined based on the solar azimuth and altitude. A region around the intersection, centered on the intersection and coinciding with the area of the PV power station to be predicted, is selected as the target region in the predicted cloud map that will affect the PV power station. The grayscale value of each pixel in the target region is extracted, and the grayscale mean of the target region is determined based on the grayscale value of each pixel. The solar irradiance loss coefficient corresponding to the cloud map with the same grayscale mean in historical data and weather data close to the weather forecast data for the predicted region is determined as the solar irradiance loss coefficient corresponding to the target region. The surface irradiance at time t3 in the future is predicted based on the solar irradiance loss coefficient corresponding to the target region. This method effectively predicts complex cloud motion processes such as deformation and formation and dissipation, more accurately predicts cloud motion trajectories, improves the accuracy of surface irradiance attenuation predictions, and ensures the safe and stable operation of the power grid system.

[0066] An embodiment of the present application provides a processor, which is used to run a program, wherein the program, when running, executes the above-mentioned method for predicting regional surface radiation intensity based on cloud movement trajectories.

[0067] An embodiment of the present application provides a storage medium having a program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting regional surface radiation intensity based on cloud movement trajectories.

[0068] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected via a system bus. The processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor A01, a method for predicting regional surface radiation intensity based on the cloud movement trajectory is implemented. The display screen A04 of the computer device can be a liquid crystal display or an electronic ink display, and the input device A05 of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0069] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0070] An embodiment of the present application provides a device, comprising a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are performed:

[0071] Obtain the current satellite cloud image of the photovoltaic power station to be predicted and the weather forecast data of the area to be predicted where the photovoltaic power station to be predicted is located, determine the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data, use the cloud displacement vector calculation model to determine the predicted displacement vector of the cloud in the area to be predicted based on the cloud movement pattern of the area to be predicted, determine the predicted cloud map of the area to be predicted based on the current position and predicted displacement vector of the cloud in the area to be predicted, the predicted cloud map includes the predicted position of the cloud in the area to be predicted, determine the target area in the predicted cloud map that has an impact on the photovoltaic power station to be predicted based on the geographical location of the photovoltaic power station to be predicted, determine the solar irradiance reduction coefficient of the target area, and determine the surface radiation intensity of the area to be predicted at a preset time in the future based on the historical measured surface radiation intensity under historical clear sky conditions in the area to be predicted at a preset time in the future and the solar irradiance reduction coefficient.

[0072] In one embodiment, determining the solar irradiance loss coefficient of the target area includes: extracting the grayscale value of each pixel in the target area; determining the grayscale mean of the target area based on the grayscale value of each pixel; and determining the solar irradiance loss coefficient corresponding to the cloud map in the historical data that has the same grayscale mean and weather data close to the weather forecast data of the area to be predicted as the solar irradiance loss coefficient corresponding to the target area.

[0073] In one embodiment, the solar irradiance loss coefficient λ corresponding to the cloud image in the historical data is calculated according to the following formula (1):

[0074]

[0075] Among them, I0 is the measured surface radiation intensity, and I1 is the measured surface radiation intensity under clear sky conditions.

[0076] In one embodiment, determining the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted based on the geographical location of the photovoltaic power station to be predicted includes: determining the solar azimuth and solar altitude angle based on the geographical location of the photovoltaic power station to be predicted; determining the intersection of the sunlight and the cloud cluster in the prediction cloud map based on the solar azimuth and solar altitude angle; and selecting an area around the intersection that is consistent with the area of the photovoltaic power station to be predicted as the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted, with the intersection as the center.

[0077] In one embodiment, determining the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data includes: obtaining historical satellite cloud image sequence images and historical weather data of the photovoltaic power station; after preprocessing the historical satellite cloud image sequence images, dividing the preprocessed historical satellite cloud image sequence images into multiple image pairs, wherein each image pair includes two sequence images at adjacent times; for each image pair, extracting the feature vector of the image pair and the weather data corresponding to the image pair; using a clustering algorithm, classifying the multiple image pairs according to the weather data corresponding to each image pair to determine the cloud movement patterns corresponding to the multiple image pairs; establishing a cloud movement pattern recognition model according to the cloud movement patterns corresponding to the multiple image pairs; and using the cloud movement pattern recognition model, determining the cloud movement pattern of the area to be predicted based on the current satellite cloud image and weather forecast data.

[0078] In one embodiment, before using a cloud displacement vector calculation model to determine the predicted displacement vector of the cloud in the to-be-predicted area according to the cloud motion pattern of the to-be-predicted area, for each image pair, the cloud displacement vector of the image pair is determined separately based on multiple cloud trajectory tracking algorithms; for each image pair, the weight combination of the image pair is determined according to the cloud displacement vectors determined by all the cloud trajectory tracking algorithms; for each cloud motion pattern, the optimal weight combination of the cloud motion pattern is determined according to the weight combination of each image pair under the cloud motion pattern, so as to establish a cloud displacement vector calculation model.

[0079] In one embodiment, for each image pair, determining the cloud displacement vector of each image pair based on multiple cloud trajectory tracking algorithms includes calculating the cloud displacement vector of each image pair according to formula (2):

[0080]

[0081] X and Y represent the displacement vector of the cloud in the image pair in the X direction and the Y direction, respectively. f1, f2, ..., f n Represents different weights, V1, V2, ..., V n represents different cloud trajectory tracking algorithms, V i (x) represents the displacement vector of the cloud in the X direction calculated by the ith cloud trajectory tracking algorithm, V i (y) represents the displacement vector of the cloud in the Y direction calculated using the i-th cloud trajectory tracking algorithm.

[0082] In one embodiment, for each cloud motion mode, the optimal weight combination of the cloud motion mode is determined according to the weight combination of each image pair under the cloud motion mode to establish a cloud displacement vector calculation model, including: determining all image pairs under each cloud motion mode, and the weight combination of each image pair; for each cloud motion mode, determining the mean of each weight in the weight combination according to the weight combination of all image pairs under the cloud motion mode; for each cloud motion mode, determining the weight combination composed of the mean of each weight as the optimal weight combination of the cloud motion mode.

[0083] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0084] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1The steps for the function specified in one or more boxes.

[0087] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0088] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0089] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0090] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0091] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for predicting regional surface radiation intensity based on cloud movement trajectories, characterized in that: The prediction method comprises: Obtaining a current satellite cloud image of the photovoltaic power station to be predicted and weather forecast data for the area to be predicted where the photovoltaic power station to be predicted is located; Determining a cloud movement pattern in the area to be predicted based on the current satellite cloud image and the weather forecast data; Determining a predicted displacement vector of a cloud in the area to be predicted based on a cloud movement pattern in the area to be predicted using a cloud displacement vector calculation model; Determining a predicted cloud map of the area to be predicted based on the current position of the cloud cluster in the area to be predicted and the predicted displacement vector, wherein the predicted cloud map includes the predicted position of the cloud cluster in the area to be predicted; determining, according to the geographical location of the photovoltaic power station to be predicted, a target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted; Determining a solar irradiance loss coefficient for the target area, wherein the solar irradiance loss coefficient is determined based on the measured surface radiation intensity and the measured surface radiation intensity under clear sky conditions; The surface radiation intensity of the area to be predicted at the preset future moment is determined based on the historical surface measured radiation intensity of the area to be predicted under historical clear sky conditions at the preset future moment and the solar irradiance depreciation coefficient.

2. The prediction method according to claim 1, characterized in that The step of determining the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted according to the geographical location of the photovoltaic power station to be predicted includes: Determining the solar azimuth angle and solar altitude angle according to the geographical location of the photovoltaic power station to be predicted; Determine the intersection of sunlight and the cloud in the predicted cloud map according to the solar azimuth angle and the solar altitude angle; With the intersection as the center, an area around the intersection that is consistent with the area of the photovoltaic power station to be predicted is selected as the target area in the prediction cloud map that has an impact on the photovoltaic power station to be predicted.

3. The prediction method according to claim 1, wherein: Determining the cloud movement pattern of the to-be-predicted area according to the current satellite cloud image and the weather forecast data includes: Obtain historical satellite cloud image sequences and historical weather data of photovoltaic power plants; After preprocessing the historical satellite cloud image sequence, the preprocessed historical satellite cloud image sequence is divided into a plurality of image pairs, wherein each image pair includes two sequential images at adjacent times; For each image pair, extracting a feature vector of the image pair and weather data corresponding to the image pair; Using a clustering algorithm, classify the plurality of image pairs according to weather data corresponding to each image pair to determine cloud movement patterns corresponding to the plurality of image pairs; Establishing a cloud movement pattern recognition model based on the cloud movement patterns corresponding to multiple image pairs; The cloud movement pattern recognition model is used to determine the cloud movement pattern of the area to be predicted based on the current satellite cloud image and the weather forecast data.

4. The prediction method according to claim 3, characterized in that The method further comprises: Before determining the predicted displacement vector of the clouds in the to-be-predicted area according to the cloud motion pattern in the to-be-predicted area using a cloud displacement vector calculation model, determining the cloud displacement vector of each image pair based on multiple cloud trajectory tracking algorithms; For each image pair, a weighted combination of the image pair is determined based on the cloud displacement vectors determined by all cloud trajectory tracking algorithms; For each cloud movement pattern, the optimal weight combination of the cloud movement pattern is determined according to the weight combination of each image pair under the cloud movement pattern, so as to establish a cloud displacement vector calculation model.

5. The prediction method according to claim 4, characterized in that For each image pair, determining the cloud displacement vector of the image pair based on multiple cloud trajectory tracking algorithms includes calculating the cloud displacement vector of each image pair according to formula (2): （2） X and Y represent the displacement vector of the cloud in the image pair in the X direction and the Y direction, respectively. Indicates different weights, , ,..., represents different cloud trajectory tracking algorithms, represents the displacement vector of the cloud in the X direction calculated by the ith cloud trajectory tracking algorithm, represents the displacement vector of the cloud in the Y direction calculated using the i-th cloud trajectory tracking algorithm.

6. The prediction method according to claim 1, characterized in that For each cloud motion pattern, determining the optimal weight combination of the cloud motion pattern according to the weight combination of each image pair under the cloud motion pattern to establish a cloud displacement vector calculation model includes: Determine all image pairs under each cloud motion mode and the weight combination of each image pair; For each cloud movement pattern, determining the mean value of each weight in the weight combination according to the weight combination of all image pairs under the cloud movement pattern; For each cloud movement pattern, a weight combination consisting of the mean of each weight is determined as the optimal weight combination of the cloud movement pattern.

7. A processor, characterized in that: The method is configured to execute the method for predicting regional surface radiation intensity based on cloud movement trajectories according to any one of claims 1 to 6.

8. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the method for predicting regional surface radiation intensity based on cloud movement trajectories according to any one of claims 1 to 6.

Images