Regional photovoltaic power generation power prediction method and device, electronic equipment and storage medium
By using geographic information and historical power data in photovoltaic power stations for grid output coefficient prediction, the difficulty of fusion of space-time related information in distributed applications in the existing technology is solved, and the accuracy of regional photovoltaic power generation power prediction is improved Sexual and economical.
Patent Information
- Application Number
- CN202410925427.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-12-06
AI Technical Summary
The existing photovoltaic power prediction methods cannot effectively integrate the space-time information between power stations in distributed applications, and the prediction accuracy is limited. Especially in complex or unstable climate conditions, it is difficult to accurately capture the output of rapidly changing weather conditions to regional photovoltaics. The impact of the regional photovoltaic power generation power is low.
By obtaining the geographical information and historical power data of each photovoltaic power station in the target area, performing grid output coefficient prediction, and calculating the predicted power generation power of the target area. Using a method based on geographic information and historical power data, the history of distributed photovoltaic power stations is integrated with the Power data avoids taking into account the influence of weather conditions and improves spatial and temporal resolution.
It realizes accurate capture of complex spatial and temporal correlations between photovoltaic power stations, improves the accuracy of regional photovoltaic power generation power prediction, simplifies the implementation process, reduces costs, and is more economical.
Smart Images

Figure CN119093317A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of photovoltaic power generation technology, and in particular to a method, device, electronic device and storage medium for predicting regional photovoltaic power generation. Background Art
[0002] With the continuous advancement of photovoltaic power generation technology, the photovoltaic industry has ushered in a rapid development momentum. However, due to its intermittent and random nature, photovoltaic power generation systems pose severe challenges to the access and absorption capacity of local distribution networks under high penetration rates. Therefore, in order to meet these challenges, some photovoltaic power prediction methods have emerged. For existing photovoltaic power prediction methods, since physics-based models or traditional statistical models are usually used to predict photovoltaic power, they are mainly aimed at single centralized new energy sites. As a result, centralized power prediction in distributed applications has the problem of being unable to integrate spatiotemporal related information between power stations, limited prediction accuracy (especially under complex or unstable climatic conditions), and insufficient spatiotemporal resolution, making it difficult to accurately capture the impact of rapidly changing weather conditions on regional photovoltaic output. As a result, the accuracy of regional photovoltaic power generation prediction is low. Summary of the invention
[0003] The present invention provides a method, device, electronic device and storage medium for predicting regional photovoltaic power generation. By predicting the grid output coefficient and calculating the power generation according to geographic information and historical power data, the historical power data of distributed photovoltaic power stations are effectively integrated without considering the influence of weather conditions on regional photovoltaic output (i.e., the use of meteorological data can be avoided). The complex spatiotemporal correlations between photovoltaic power stations are accurately captured, the spatiotemporal resolution is improved, and the accuracy of regional photovoltaic power generation prediction is thereby improved.
[0004] The first aspect of the present disclosure provides a method for predicting regional photovoltaic power generation, comprising:
[0005] Obtain geographic information and historical power data of each photovoltaic power station in the target area;
[0006] Performing grid output coefficient prediction based on the geographic information and the historical power data to obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate the photovoltaic power generation performance at the prediction time;
[0007] The predicted power generation power of the target area is calculated based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0008] A second aspect of the present disclosure provides a regional photovoltaic power generation power prediction device, comprising:
[0009] An acquisition module is used to obtain geographic information and historical power data of each photovoltaic power station in the target area;
[0010] A prediction module, configured to perform grid output coefficient prediction based on the geographic information and the historical power data, and obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate the photovoltaic power generation performance at the prediction moment;
[0011] The calculation module is used to calculate the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0012] A third aspect of the present disclosure provides an electronic device, which includes a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the above-mentioned regional photovoltaic power generation power prediction method.
[0013] A fourth aspect of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the above-mentioned regional photovoltaic power generation power prediction method.
[0014] In the technical solution provided by the present disclosure, the geographic information and historical power data of each photovoltaic power station in the target area are obtained; a grid output coefficient prediction is performed based on the geographic information and the historical power data to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate the photovoltaic power generation performance at the prediction moment; and the predicted power generation power of the target area is calculated based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit. In the embodiment of the present disclosure, by performing grid output coefficient prediction and power generation calculation based on geographic information and historical power data, the historical power data of the distributed photovoltaic power station is effectively integrated, and there is no need to consider the factors affecting the regional photovoltaic output under weather conditions (i.e., the use of meteorological data can be avoided), and the complex spatiotemporal correlations between photovoltaic power stations are accurately captured, the spatiotemporal resolution is improved, and the accuracy of regional photovoltaic power generation power prediction is improved, and it is easy to implement and more economical. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A schematic diagram of an embodiment of a method for predicting regional photovoltaic power generation in an embodiment of the present disclosure;
[0016] Figure 2 It is a schematic diagram of another embodiment of the method for predicting regional photovoltaic power generation in the embodiment of the present disclosure;
[0017] Figure 3 It is a schematic diagram of an embodiment of the distribution status of photovoltaic power stations in a grid in an embodiment of the present disclosure;
[0018] Figure 4 A schematic diagram of an embodiment of a device for predicting regional photovoltaic power generation in an embodiment of the present disclosure;
[0019] Figure 5 Schematic diagram of an electronic device in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0020] The embodiments of the present disclosure provide a method, device, electronic device and storage medium for predicting regional photovoltaic power generation, which improve the accuracy of regional photovoltaic power generation prediction.
[0021] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present disclosure 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 interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "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.
[0022] With the continuous advancement of photovoltaic power generation technology, the photovoltaic industry has ushered in a rapid development momentum. In particular, distributed photovoltaic power generation. In recent years, the newly installed capacity of distributed photovoltaic power generation has exceeded the newly installed capacity of centralized photovoltaic power generation, becoming the main force driving the growth of new photovoltaic installed capacity. However, due to its intermittent and random nature, distributed photovoltaic power generation systems pose severe challenges to the access and absorption capacity of local distribution networks under high penetration rates. These challenges are mainly reflected in the fact that the grid connection of a large number of distributed photovoltaics has changed the operating characteristics of the regional power grid and the source-grid interaction mechanism, increased the uncertainty in the system's power generation and consumption balance and real-time scheduling process, and increased the cost of full absorption, thereby affecting the safety, stability and economy of power grid operation.
[0023] In order to meet these challenges, it is particularly important to carry out multi-temporal and spatial scale, high-precision photovoltaic power generation forecasting, especially distributed photovoltaic output forecasting. Distributed photovoltaic output forecasting not only helps to optimize the grid dispatching strategy and ensure the stability of the grid, but also promotes the large-scale access and efficient utilization of renewable energy and ensures the stable operation of the grid. Existing photovoltaic power forecasting methods are mainly aimed at single centralized new energy sites, and the methods usually include physics-based models or traditional statistical models. However, centralized power forecasting methods show obvious deficiencies in distributed applications, such as: the inability to integrate spatiotemporal related information between power stations, limited prediction accuracy, especially under complex or unstable climatic conditions, and insufficient spatiotemporal resolution, making it difficult to accurately capture the impact of rapidly changing weather conditions on regional photovoltaic output. As a result, the accuracy of regional photovoltaic power forecasting is relatively low.
[0024] Based on this, the embodiments of the present disclosure provide a method, device, electronic device and storage medium for predicting regional photovoltaic power generation, which are mainly used in photovoltaic power generation.
[0025] It is understandable that the execution subject of the present disclosure may be a regional photovoltaic power generation prediction device, or a terminal or a server, which is not limited here. The present disclosure embodiment is described by taking a server as the execution subject as an example.
[0026] For ease of understanding, the specific process of the embodiment of the present disclosure is described below. Figure 1 , an embodiment of the method for predicting regional photovoltaic power generation in the embodiment of the present disclosure includes:
[0027] 101. Obtain geographic information and historical power data of each photovoltaic power station in the target area;
[0028] Among them, the target area can be determined according to the demand for output coefficient prediction. For example, the target area can be the area corresponding to the entire power grid, or it can be the area where the line requiring photovoltaic output coefficient prediction is located (for example, the town area, urban area); the target area can be understood as the area corresponding to the photovoltaic output coefficient prediction demand and the density of photovoltaic power stations is greater than the preset density, that is, the target area is the designated area corresponding to the photovoltaic output coefficient prediction demand, and the density of photovoltaic power stations in the target area is greater than the preset density.
[0029] The photovoltaic power station may be a distributed photovoltaic power station or a household photovoltaic power station or other, which is not limited here. The geographical information may include longitude and latitude information. The historical power data may be the actual power generation during a preset period before the current time, for example, the actual power generation during the past month.
[0030] As an example but not limitation, the geographic information and historical power data of each photovoltaic power station in the target area may be extracted from a database corresponding to the monitoring system of the photovoltaic power station or a geographic information system database.
[0031] 102. Perform grid output coefficient prediction based on geographic information and historical power data to obtain predicted photovoltaic output coefficients of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate photovoltaic power generation performance at the prediction time;
[0032] As an example but not limitation, the prediction of gridded output coefficient based on geographic information and historical power data to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit can be understood as: generating a gridded historical photovoltaic output coefficient image based on geographic and historical power data; predicting based on the historical photovoltaic output coefficient image to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit. Specifically, it can be as follows: gridding the target area to obtain multiple grid cells corresponding to the target area; generating a gridded historical photovoltaic output coefficient image based on the geographic information and historical power data of multiple grid cells, the historical photovoltaic output coefficient image including the historical output coefficient corresponding to each grid cell; predicting based on the historical photovoltaic output coefficient image (the historical output coefficient corresponding to each grid cell) to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0033] 103. Calculate the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0034] As an example but not limitation, the predicted power generation of each photovoltaic power station in the grid unit is predicted based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit; the predicted power generation of all photovoltaic power stations is integrated to obtain the predicted power generation of the target area.
[0035] In the disclosed embodiment, by predicting the grid output coefficient and calculating the generated power based on geographic information and historical power data, the historical power data of distributed photovoltaic power stations is effectively integrated, and there is no need to consider the impact of weather conditions on regional photovoltaic output (i.e., the use of meteorological data can be avoided). The complex spatiotemporal correlations between photovoltaic power stations are accurately captured, the spatiotemporal resolution is improved, and the accuracy of regional photovoltaic power generation prediction is thereby improved. The method is easy to implement and more economical.
[0036] See also Figure 2 Another embodiment of the method for predicting regional photovoltaic power generation in the embodiment of the present disclosure includes:
[0037] 201. Obtain geographic information and historical power data of each photovoltaic power station in the target area;
[0038] The execution process of step 201 is similar to the execution process of step 101, and will not be repeated here.
[0039] 202. Determine a historical output coefficient of each photovoltaic power station based on historical power data of each photovoltaic power station; wherein the historical output coefficient is used to indicate photovoltaic power generation performance at each historical moment in a historical period;
[0040] The historical output coefficient can be understood as the ratio of the actual power to the theoretical maximum power generation at each historical moment in the historical period. As an example but not a limitation, the historical output coefficient can also be used to indicate the photovoltaic power generation performance at each historical moment in a day in the historical period. The historical output coefficient of each photovoltaic power station is used to indicate the photovoltaic power generation performance of the photovoltaic power station at the historical moment and the location of the photovoltaic power station.
[0041] As an example but not limitation, data preprocessing is performed on the historical power data of each photovoltaic power station to obtain the preprocessed power data of each photovoltaic power station; all power data corresponding to the same time in the preprocessed power data of each photovoltaic power station are obtained to obtain the power data of each photovoltaic power station at each time, wherein the time refers to each time within a day; based on the power data of each photovoltaic power station at each time, the theoretical maximum power at each time is determined; the power data of each photovoltaic power station at each time is averaged to obtain the target actual power of each photovoltaic power station at each time; based on the target actual power and the theoretical maximum power of each photovoltaic power station at each time, calculation is performed to obtain the historical output coefficient of each photovoltaic power station at each time; the historical output coefficient of each photovoltaic power station at all times is determined as the historical output coefficient of each photovoltaic power station.
[0042] In one implementation, in determining the historical output coefficient of each photovoltaic power station based on the historical power data of each photovoltaic power station, the following can be done: preprocessing the historical power data of each photovoltaic power station to obtain the preprocessed historical power data of each photovoltaic power station; sorting the power data at each target moment in the preprocessed historical power data of each photovoltaic power station to obtain the sorted power data at each target moment of each photovoltaic power station, wherein the target moment is used to indicate the same moment of each day in the historical period; determining the theoretical maximum power at each target moment of each photovoltaic power station based on the sorted power data at each target moment of each photovoltaic power station; and calculating the historical output coefficient of each photovoltaic power station based on the actual power value and the theoretical maximum power at each target moment of each photovoltaic power station.
[0043] As an example but not limitation, outliers are eliminated from the historical power data of each photovoltaic power station to obtain the eliminated data of each photovoltaic power station; missing data in the eliminated data of each photovoltaic power station and historical power data of adjacent photovoltaic power stations corresponding to the missing data are obtained; and the missing data are interpolated based on the historical power data of adjacent photovoltaic power stations corresponding to the missing data to obtain the pre-processed historical power data of each photovoltaic power station.
[0044] The target time is, for example, 10-11 a.m. every day in the historical period. When determining the theoretical maximum power of each photovoltaic power station at each target time based on the sorted power data at each target time of each photovoltaic power station, specifically, the pre-processed historical power data of all photovoltaic power stations can be classified according to the target time to obtain the power data at each target time; the power data at each target time can be sorted in order from small to large values to obtain the sorted power data at each target time of each photovoltaic power station; the power ranked in a preset position in the sorted power data at each target time of each photovoltaic power station is determined as the theoretical maximum power, and the theoretical maximum power of each photovoltaic power station at each target time is obtained, wherein the preset position is, for example, ranked as the 95th. The theoretical maximum power of all target times of each photovoltaic power station can constitute a theoretical maximum power curve.
[0045] The actual power value of each photovoltaic power station at each target time is divided by the theoretical maximum power to obtain the historical output coefficient of each photovoltaic power station.
[0046] By cleaning the historical power data of the photovoltaic power station, calculating the theoretical maximum power and then calculating the historical output coefficient, the data quality of the theoretical maximum power calculation is improved, and the accuracy and reliability of the historical output coefficient of each photovoltaic power station are improved.
[0047] 203. Divide the target area into grids to obtain a plurality of grid units corresponding to the target area;
[0048] Among them, each grid unit (grid pixel) corresponds to an area in the geographic space, which can be simply understood as each grid unit (grid pixel) does not correspond to a point, but to a surface (a small area).
[0049] As an example but not limitation, edge power station information of a target area is obtained; a grid boundary is determined based on the target area and the edge power station information of the target area, wherein the size of an area formed by the grid boundary is less than or equal to the size of the target area, that is, when the grid boundary corresponds to the edge of the target area, the size of the area formed by the grid boundary is equal to the size of the target area; a predefined grid size and grid resolution are obtained, and the area formed by the grid boundary is grid-divided based on the grid size and the grid resolution to obtain a plurality of grid units corresponding to the target area.
[0050] In one implementation, when a target area is divided into a grid and a plurality of grid cells corresponding to the target area are obtained, the following can be done: the spatial range and resolution of the grid are determined based on effective power station information, wherein the effective power station information is used to indicate the spatial distribution information and required analysis accuracy of a preset number of photovoltaic power stations that can be determined within a preset distance range; based on the spatial range and resolution, a grid corresponding to the target area is created, wherein the grid includes a plurality of grid cells of uniform size, and each grid cell corresponds to an area in geographic space; based on the location of the photovoltaic power station at the edge of the target area, the boundary of the grid is adjusted to obtain a plurality of grid cells corresponding to the target area.
[0051] Among them, the spatial range can be understood as the area range that constitutes the overall grid, and the resolution can be understood as the size (dimension) of the grid unit. Generally speaking, for uniformity and standardization, as well as for ease of calculation and operation, the plane shape of the grid unit is a square.
[0052] As an example but not limitation, the number of first photovoltaic power stations within a preset distance range is determined based on the geographic information of each photovoltaic power station in the target area; if the number of first photovoltaic power stations is greater than or equal to a preset number threshold, the distribution of photovoltaic power stations within the preset distance range is determined as the spatial distribution information of effective power stations, and the required analysis accuracy of the effective power stations is determined based on the preset distance range, wherein the required analysis accuracy is the resolution of the grid unit, and the required analysis accuracy is less than the preset distance range; if the number of first photovoltaic power stations is less than the preset number threshold, the preset distance range is adjusted based on a preset step length to obtain an adjusted distance range, wherein the preset step length may be less than or equal to the preset distance range; the second photovoltaic power station within the adjusted distance range is determined based on the geographic information of each photovoltaic power station in the target area The number of photovoltaic power stations; if the second number of photovoltaic power stations is greater than or equal to the preset number threshold, the distribution of photovoltaic power stations within the adjusted distance range is determined as the spatial distribution information of the effective power stations, and the required analysis accuracy of the effective power stations is determined based on the adjusted distance range; if the second number of photovoltaic power stations is less than the preset number threshold, the adjusted distance range is adjusted based on a preset step size until the number of photovoltaic power stations within the distance range after the second adjustment is greater than or equal to the preset number threshold, the distribution of photovoltaic power stations within the distance range after the second adjustment is determined as the spatial distribution information of the effective power stations, and the required analysis accuracy of the effective power stations is determined based on the distance range after the second adjustment, wherein the distance range after the second adjustment is less than the preset distance range threshold, and the preset distance range threshold is, for example, three kilometers.
[0053] As an example but not limitation, the distribution information of photovoltaic power stations at the edge of the target area is obtained, wherein the distribution information of photovoltaic power stations includes but is not limited to the number of photovoltaic power stations and the geographical information of their locations; if the distribution information of photovoltaic power stations indicates that there is no photovoltaic power station, the location of the target photovoltaic power station is obtained, wherein the location of the target photovoltaic power station is the location of the photovoltaic power station in the target area that is closest to the edge of the target area; the boundary of the grid is adjusted to a position at a specified distance from the location of the target photovoltaic power station to obtain an adjusted grid, i.e., a plurality of grid cells corresponding to the target area; if the distribution information of photovoltaic power stations indicates that there is a photovoltaic power station, the boundary of the grid is not adjusted, and all grid cells in the created grid are determined as the plurality of grid cells corresponding to the target area.
[0054] By gridding the spatial range and resolution of the grid determined based on the effective power station information and adjusting the grid boundaries, precise grid spatial range definition and accurate spatial data analysis are provided, the calculation effect is optimized, and the spatial data of photovoltaic power stations in the target area can be processed more accurately, reliably and efficiently.
[0055] 204. Generate a historical photovoltaic output coefficient image based on the geographic information and historical output coefficient of each photovoltaic power station and a plurality of grid cells;
[0056] Among them, there may be at least one photovoltaic power station distributed in each grid unit, and each photovoltaic power station corresponds to a historical output coefficient. There may also be no photovoltaic power station distributed in each grid unit; each grid unit corresponds to a historical grid output coefficient, and the historical grid output coefficient can be understood as the output coefficient obtained after comprehensive processing of the historical output coefficients of the photovoltaic power station in each grid unit.
[0057] As an example but not limitation, based on the geographic information of each photovoltaic power station, the grid unit in which each photovoltaic power station is located in multiple grid units is determined to obtain the distribution status of the photovoltaic power stations in each grid unit; the historical grid output coefficient of each grid unit is determined according to the distribution status, geographic information and historical output coefficient of the photovoltaic power stations in each grid unit; and a historical photovoltaic output coefficient image is generated based on all grid cells and the historical grid output coefficients of all grid cells.
[0058] In one implementation, when generating a historical photovoltaic output coefficient image based on the geographic information and historical output coefficient of each photovoltaic power station and multiple grid units, the photovoltaic power station distributed in each grid unit among the multiple grid units can be determined based on the geographic information of each photovoltaic power station; the historical grid output coefficient of each grid unit can be determined according to the historical output coefficient of the photovoltaic power station distributed in each grid unit; and the historical photovoltaic output coefficient image can be generated based on the multiple grid units and the historical grid output coefficient of each grid unit.
[0059] The distribution of photovoltaic power stations in the grid can be shown as follows: Figure 3 As shown, Figure 3 The squares in correspond to grid cells. Figure 3 Each point in corresponds to a photovoltaic power station.
[0060] As an example but not limitation, the geographic information of each photovoltaic power station can be converted into corresponding grid coordinates; based on the converted grid coordinates of each photovoltaic power station, the grid unit in which each photovoltaic power station is located among multiple grid units is determined; and the photovoltaic power stations in each grid unit are counted to obtain the photovoltaic power stations distributed in each grid unit.
[0061] The historical photovoltaic output coefficient image is used to indicate the spatial distribution of the historical grid output coefficients of all photovoltaic power stations in the target area.
[0062] By generating a historical photovoltaic output coefficient image based on the gridded historical output coefficient of each photovoltaic power station's geographic information (i.e., the historical grid output coefficient), the spatial distribution of the photovoltaic power station output coefficient is visualized, which facilitates data analysis and comparison of the photovoltaic power station's power generation performance, and is beneficial to the efficiency, accuracy, and reliability of subsequent photovoltaic output coefficient predictions.
[0063] In one implementation, when determining the historical grid output coefficient of each grid unit according to the historical output coefficient of the photovoltaic power station distributed in each grid unit, the following can be done: if the multiple grid units include first-type grid units and second-type grid units, then determining the historical grid output coefficient of each grid unit in the first-type grid units based on all historical output coefficients of each grid unit in the first-type grid units, wherein the first-type grid units are grid units in which at least one photovoltaic power station is distributed, and the second-type grid units are grid units in which no photovoltaic power station is distributed; obtaining the distribution density of the photovoltaic power stations in the target area; and performing grid mapping on each grid unit in the second-type grid units based on the geographic information and the distribution density to obtain the historical grid output coefficient of each grid unit in the second-type grid units.
[0064] As an example but not limitation, when determining the historical grid output coefficient of each grid unit in the first type of grid unit based on all historical output coefficients of each grid unit in the first type of grid unit, specifically, if the number of photovoltaic power stations distributed in the grid units in the first type of grid unit is one, the historical output coefficient of the corresponding grid unit is determined as the historical grid output coefficient of the corresponding grid unit in the first type of grid unit; if the number of photovoltaic power stations distributed in the grid units in the first type of grid unit is more than one, all historical output coefficients of the corresponding grid unit are calculated based on a preset algorithm to obtain the historical grid output coefficient of the corresponding grid unit in the first type of grid unit, wherein, as an example but not limitation, the preset algorithm can be an average value algorithm, a minimum value algorithm, a maximum value algorithm or an interpolation algorithm. By processing all historical output coefficients of the corresponding grid unit with the preset algorithm, the accuracy and reliability of the historical grid output coefficient of the grid unit are improved.
[0065] By determining the historical grid output coefficient of the grid unit according to the spatial distribution of photovoltaic power stations in the grid unit and the historical output coefficient of the distribution, it can better help capture the correlation of the spatial distribution of the grid output coefficient, improve the accuracy and reliability of the historical grid output coefficient of each grid unit, and help improve the accuracy of regional photovoltaic power generation prediction.
[0066] In one implementation, when grid mapping is performed on each grid cell in the second type of grid cells based on geographic information and distribution density to obtain a historical grid output coefficient of each grid cell in the second type of grid cells, it is possible to: if the distribution density is greater than a preset value, determine a target grid cell corresponding to each grid cell in the second type of grid cells based on the geographic information, wherein the target grid cell is at least one adjacent grid cell; and interpolate the historical output coefficient or the historical grid output coefficient in the target grid cell to obtain a historical grid output coefficient of each grid cell in the second type of grid cells.
[0067] The geographic information can be understood as the location information of the photovoltaic power station in the specific range of each grid unit in the second type of grid unit in the geographic space. The distribution density can be understood as the distribution density of the photovoltaic power station in the first type of grid unit, and can also be understood as the distribution density of the photovoltaic power station in the grid unit in the specific range of each grid unit in the second type of grid unit.
[0068] As an example but not a limitation, the target grid unit is the closest grid unit with adjacent photovoltaic power stations. When determining the target grid unit corresponding to each grid unit in the second type of grid unit based on geographic information, specifically, the following can be done: determine the photovoltaic power station within the preset range of each grid unit in the second type of grid unit; perform distance calculation based on the geographic information of the photovoltaic power station within the preset range and the coordinates of the center point of each grid unit in the second type of grid unit to obtain the distance between each photovoltaic power station within the preset range and the corresponding grid unit in the second type of grid unit; determine the grid unit corresponding to the photovoltaic power station with the smallest distance value as the target grid unit. By determining the grid unit corresponding to the photovoltaic power station with the smallest distance value as the target grid unit, the accuracy of determining the target grid unit can be improved in terms of the correlation of spatial distribution.
[0069] As an example but not limitation, the target grid unit is the adjacent grid unit with the most photovoltaic power stations. When determining the target grid unit corresponding to each grid unit in the second type of grid unit based on geographic information, specifically, the following can be done: determine the photovoltaic power station within a preset range of each grid unit in the second type of grid unit; count the grid unit with the largest number of photovoltaic power stations based on the photovoltaic power stations within the preset range; and determine the grid unit with the largest number of photovoltaic power stations as the target grid unit. By determining the grid unit with the largest number of photovoltaic power stations as the target grid unit, the accuracy of determining the target grid unit can be improved in terms of quantity distribution.
[0070] As an example but not a limitation, the target grid unit is at least one adjacent grid unit, and the photovoltaic power station within the preset range of each grid unit in the second type of grid unit is determined, and the grid unit corresponding to the photovoltaic power station within the preset range is determined as the target grid unit; or the grid units with historical grid output coefficients distributed within the preset range of the second type of grid units are determined among all grid units, and the grid units with historical grid output coefficients distributed within the preset range are determined as the target grid units. When the historical output coefficient or historical grid output coefficient in the target grid unit is interpolated to obtain the historical grid output coefficient of each grid unit in the second type of grid unit, specifically, it can be: interpolated to obtain the historical grid output coefficient of each grid unit in the second type of grid unit; or the historical output coefficient of each grid unit in the target grid unit is averaged to obtain the average of each grid unit in the target grid unit, and the average of all grid units in the target grid unit is interpolated to obtain the historical grid output coefficient of each grid unit in the second type of grid unit. By interpolating the photovoltaic power plants within a preset range, the error in the correlation of spatial distribution can be reduced to a large extent, and the accuracy of determining the target grid unit can be improved.
[0071] By interpolating the output coefficient of at least one adjacent grid cell when the distribution density is greater than a preset value, the historical grid output coefficient of the grid cell that is not directly covered by the historical output coefficient is obtained, thereby filling the data gaps, improving the integrity and continuity of the historical grid output coefficient in the grid, enhancing the spatial resolution of the historical grid output coefficient in the target area, fully mining and utilizing the massive data resources in the photovoltaic power grid, and generating high-precision historical grid output coefficients.
[0072] In one implementation, when grid mapping is performed on each grid cell in the second type of grid cells based on geographic information and distribution density to obtain a historical grid output coefficient of each grid cell in the second type of grid cells, the following can be done: if the distribution density is less than or equal to a preset value, terrain data corresponding to a first specific grid cell among multiple grid cells is obtained; spatial distribution characteristics are obtained from the terrain data corresponding to the first specific grid cell and the specific output coefficient of the second specific grid cell in the first type of grid cells through a preset deep learning model, wherein the specific output coefficient is a historical output coefficient or a historical grid output coefficient; and the historical grid output coefficient of each grid cell in the second type of grid cells is determined based on the spatial distribution characteristics.
[0073] Among them, the first specific grid unit may be all grid units among the multiple grid units, or a grid unit in the multiple grid units that is within a preset range from each grid unit in the second type of grid unit. The second specific grid unit may be all grid units in the first type of grid units, or a grid unit in the first type of grid units that is within a preset range from each grid unit in the second type of grid unit. The terrain data may include altitude, and the terrain data may also include terrain contours, slopes, and directions. As an example but not limitation, the preset deep learning model may be a deep learning algorithm such as a convolutional neural network (CNN) or a generative adversarial network (GAN) or a diffusion model that can learn spatial distribution characteristics from specific output coefficients and terrain data. Spatial distribution characteristics include but are not limited to the characteristics of the spatial distribution of historical grid output coefficients, the characteristics of the spatial distribution of historical output coefficients, and the spatial characteristics of the distribution of photovoltaic power stations on the terrain.
[0074] As an example but not limitation, the output coefficient of the grid unit is calculated based on the spatial distribution characteristics through a deep learning model or a preset calculation model to obtain the historical grid output coefficient of each grid unit in the second type of grid units.
[0075] By effectively integrating the historical power data of photovoltaic power stations through a deep learning model when the distribution density is less than or equal to the preset value, the complex spatiotemporal correlation between power stations is accurately captured, the accuracy and reliability of the historical grid output coefficient of the target area are improved, and the historical grid output coefficient in a large area can be quickly inferred and predicted, which helps to improve the prediction accuracy and prediction range, making power grid dispatch more efficient and reliable, and helping to ensure that power grid operators, power generation groups or third-party operating companies can obtain accurate prediction results in the shortest time and effectively respond to power grid load changes and market demand. In addition, it also helps to reduce the implementation cost of prediction technology, which not only reduces the initial investment, but also reduces maintenance costs, making regional photovoltaic power generation power prediction technology more economical and easy to implement.
[0076] 205. Predicting based on the historical photovoltaic output coefficient image by using a preset target output coefficient prediction model to obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit;
[0077] Among them, the target output coefficient prediction model is a model used to capture the output changes in the time series of historical photovoltaic output coefficient images, long-term dependencies and complex patterns in spatial series data.
[0078] As an example but not a limitation, before the preset target output coefficient prediction model is used to predict based on the historical photovoltaic output coefficient image to obtain the predicted output coefficient spatial distribution information, the following will be done: construct an initial output coefficient prediction model, wherein the core framework in the initial output coefficient prediction model is an architecture for capturing long-term dependencies and complex patterns in the spatial sequence data in the photovoltaic output coefficient image, for example, a Transformer architecture or a Vision Transformer (ViT) architecture; obtain a historical photovoltaic output coefficient image, wherein the historical photovoltaic output coefficient image is a photovoltaic output coefficient image generated before the time of the historical power data of each photovoltaic power station in the target area; train the initial output coefficient prediction model through the historical photovoltaic output coefficient image until the loss function value converges to obtain the target output coefficient prediction model. In the training process, the initial output coefficient prediction model can learn the output changes of the photovoltaic power station in the time series and the correlation in the spatial distribution, and can capture the spatial dependency of the output coefficient in a large range through the multi-head attention mechanism. The output changes of the photovoltaic power station in the time series can be understood as the fluctuation of the power generation capacity or power output of the photovoltaic power station at different time points.
[0079] By combining data gridding and geographic information, irregular power station power data is generated into high-precision gridded output coefficient data (historical photovoltaic output coefficient image), and the spatial distribution of output coefficient at future moments is predicted through the target output coefficient prediction model combined with the gridded output coefficient data, thereby improving the accuracy of regional photovoltaic output coefficient prediction.
[0080] In one implementation, the historical photovoltaic output coefficient image is analyzed by a preset target output coefficient prediction model to obtain output space information, wherein the output space information is used to indicate the change of the output coefficient in the time series and the correlation in the spatial distribution; the target output coefficient prediction model is used to make a prediction based on the output space information and the prediction time to obtain the predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit.
[0081] The historical photovoltaic output coefficient image is used as the input of the preset target output coefficient prediction model, and the output of the target output coefficient prediction model is the photovoltaic output coefficient image corresponding to the prediction time, that is, the predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit. Among them, the output space information includes but is not limited to the photovoltaic output information and output mode of different areas in the target area.
[0082] As an example but not limitation, high-dimensional feature representations in historical photovoltaic output coefficient images are extracted through a multi-head attention mechanism in a preset target output coefficient prediction model to obtain global features; global features are converted into spatial distribution features through a decoder in the target output coefficient prediction model to obtain output space information, or, through an analysis algorithm in the target output coefficient prediction model, global features are analyzed to obtain output space information, wherein the analysis algorithm may be a clustering algorithm, a dimensionality reduction algorithm, a principal component analysis algorithm, or the like.
[0083] Among them, the prediction time can be understood as the time corresponding to the historical photovoltaic output coefficient image plus the target prediction time. The predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit is the photovoltaic output coefficient corresponding to the target prediction time of each photovoltaic power station in each grid unit on the basis of the time corresponding to the historical photovoltaic output coefficient image in the target area, wherein the target prediction time can be understood as T0+n*ΔT, T0 is the time corresponding to the historical photovoltaic output coefficient image, and n is the number of loop iterations. For example, the target output coefficient prediction model predicts the output coefficient spatial distribution information of the next moment T0+ΔT based on the historical output coefficient image of the current moment (i.e., the moment corresponding to the historical photovoltaic output coefficient image) T0. Through loop iteration, the output coefficient spatial distribution information of the target prediction time T0+n*ΔT can be obtained, that is, the predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit.
[0084] By adopting a deep learning architecture that captures the long-term dependencies and complex patterns in the spatial sequence data in the photovoltaic output coefficient image in the target output coefficient prediction model, it is possible to accurately predict the spatial distribution of the output coefficient at a future moment (i.e., the prediction time) (i.e., the predicted output coefficient spatial distribution information), thereby achieving high-accuracy and high-resolution regional photovoltaic output coefficient prediction, thereby improving the accuracy of regional photovoltaic power generation prediction.
[0085] 206. Calculate the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0086] In one implementation, power generation calculation parameters of each photovoltaic power station in a grid unit are obtained, wherein the power generation calculation parameters are installed capacity and / or theoretical maximum power; predicted power generation power of each photovoltaic power station is calculated based on predicted photovoltaic output coefficient and power generation calculation parameters of each photovoltaic power station in the grid unit; predicted power generation power of all photovoltaic power stations in a target area is integrated to obtain predicted power generation power of the target area.
[0087] Among them, under normal circumstances (when no fault occurs), the installed capacity and the theoretical maximum power are equal, but in the event of a fault (for example, a partial component or inverter failure), the theoretical maximum power may be less than the installed capacity. As an example but not a limitation, when obtaining the power generation calculation parameters of each photovoltaic power station in the grid unit, the current fault analysis information of each photovoltaic power station in the target area can be obtained; if the current fault analysis information indicates a target device failure, the installed capacity of the corresponding photovoltaic power station is obtained, and the power generation calculation parameters of the corresponding photovoltaic power station are obtained, wherein the target device failure can be understood as a (partial or complete) photovoltaic component failure or inverter failure; if the current fault analysis information indicates no fault, the theoretical maximum power of the corresponding photovoltaic power station is obtained, and the power generation calculation parameters of the corresponding photovoltaic power station are obtained.
[0088] As an example but not limitation, when obtaining the power generation calculation parameters of each photovoltaic power station in the grid unit, the installed capacity and theoretical maximum power of each photovoltaic power station in the target area can be directly obtained to obtain the power generation calculation parameters of the corresponding photovoltaic power station.
[0089] As an example but not limitation, when calculating the predicted power generation power of each photovoltaic power station based on the predicted photovoltaic output coefficient and power generation power calculation parameters of each photovoltaic power station in the grid unit, it can be: if the power generation power calculation parameter of the photovoltaic power station in the grid unit is the installed capacity, then the product of the predicted photovoltaic output coefficient and the installed capacity of the corresponding photovoltaic power station is calculated to obtain the predicted power generation power of the corresponding photovoltaic power station; if the power generation power calculation parameter of the photovoltaic power station in the grid unit is the theoretical maximum power, then the product of the predicted photovoltaic output coefficient and the theoretical maximum power of the corresponding photovoltaic power station is calculated to obtain the predicted power generation power of the corresponding photovoltaic power station; if the power generation power calculation parameter of each photovoltaic power station in the grid unit is the installed capacity and the theoretical maximum power, then the product of the predicted photovoltaic output coefficient and the installed capacity of each photovoltaic power station is calculated to obtain the first power generation power of each photovoltaic power station, the predicted photovoltaic output coefficient and the theoretical maximum power of each photovoltaic power station are calculated to obtain the second power generation power of each photovoltaic power station, and the arithmetic mean of the first power generation power and the second power generation power of each photovoltaic power station is calculated to obtain the predicted power generation power of each photovoltaic power station.
[0090] When the predicted power generation of all photovoltaic power stations in the target area is integrated to obtain the predicted power generation of the target area, the predicted power generation of all photovoltaic power stations in the target area may be added together to obtain the predicted power generation of the target area.
[0091] By obtaining the corresponding target predicted output coefficient from the spatial distribution information of the predicted output coefficient based on the predicted demand for photovoltaic power generation, and calculating it in combination with the installed capacity and / or theoretical maximum power, the actual power generation of the regional photovoltaic power station in a specific time period in the future can be predicted more accurately, thereby improving the effectiveness and utilization rate of the predicted regional photovoltaic output coefficient, improving the accuracy of the regional photovoltaic power generation power prediction, ensuring the reliable operation, efficiency and stability of the power grid, and improving the operating efficiency and cost of the regional photovoltaic power station.
[0092] In the disclosed embodiment, irregular power station power data is combined with data gridding and geographic information to generate high-precision gridded output coefficient data (historical photovoltaic output coefficient image), and the spatial distribution of the output coefficient at future times is predicted by combining the gridded output coefficient data with the target output coefficient prediction model, thereby realizing the visualization of the spatial distribution of the photovoltaic power station output coefficient, facilitating data analysis and data comparison of the photovoltaic power station power generation performance, and facilitating the efficiency, accuracy and reliability of subsequent photovoltaic output coefficient predictions. In addition, by adopting a deep learning architecture that captures long-term dependencies and complex patterns in spatial sequence data in the photovoltaic output coefficient image in the target output coefficient prediction model, the spatial distribution of the output coefficient at future times (i.e., the prediction time) can be accurately predicted (i.e., the predicted output coefficient spatial distribution information), thereby realizing high-accuracy and high-resolution regional photovoltaic power generation power prediction.
[0093] The above describes the method for predicting regional photovoltaic power generation in the embodiment of the present disclosure. The following describes the device for predicting regional photovoltaic power generation in the embodiment of the present disclosure. Figure 4 , an embodiment of the regional photovoltaic power generation prediction device in the embodiment of the present disclosure includes:
[0094] An acquisition module 401 is used to acquire geographic information and historical power data of each photovoltaic power station in the target area;
[0095] Prediction module 402, used to predict grid output coefficient based on geographic information and historical power data, and obtain predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate photovoltaic power generation performance at the prediction time;
[0096] The calculation module 403 is used to calculate the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0097] In the disclosed embodiment, by predicting the grid output coefficient and calculating the generated power based on geographic information and historical power data, the historical power data of distributed photovoltaic power stations is effectively integrated, and there is no need to consider the impact of weather conditions on regional photovoltaic output (i.e., the use of meteorological data can be avoided). The complex spatiotemporal correlations between photovoltaic power stations are accurately captured, the spatiotemporal resolution is improved, and the accuracy of regional photovoltaic power generation prediction is thereby improved. The method is easy to implement and more economical.
[0098] Optionally, the calculation module 403 may also be specifically used for:
[0099] Obtaining power generation calculation parameters of each photovoltaic power station in the grid unit, wherein the power generation calculation parameters are installed capacity and / or theoretical maximum power;
[0100] Calculate the predicted power generation of each photovoltaic power station based on the predicted photovoltaic output coefficient and power generation calculation parameters of each photovoltaic power station in the grid unit;
[0101] The predicted power generation of all photovoltaic power stations in the target area is integrated to obtain the predicted power generation of the target area.
[0102] Optionally, the prediction module 402 may also be specifically used for:
[0103] Determine a historical output coefficient of each photovoltaic power station based on historical power data of each photovoltaic power station; wherein the historical output coefficient is used to indicate the photovoltaic power generation performance at each historical moment in the historical period;
[0104] Divide the target area into grids to obtain multiple grid cells corresponding to the target area;
[0105] Generate a historical photovoltaic output coefficient image based on the geographic information and historical output coefficient of each photovoltaic power station and multiple grid cells;
[0106] The preset target output coefficient prediction model is used to predict the historical photovoltaic output coefficient image to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
[0107] Optionally, the prediction module 402 may also be specifically used for:
[0108] Determining a photovoltaic power station distributed in each grid cell of a plurality of grid cells based on geographic information of each photovoltaic power station;
[0109] Determine the historical grid output coefficient of each grid unit according to the historical output coefficient of the photovoltaic power station distributed in each grid unit;
[0110] A historical photovoltaic output coefficient image is generated based on a plurality of grid cells and a historical grid output coefficient of each grid cell.
[0111] Optionally, the prediction module 402 may also be specifically used for:
[0112] If the plurality of grid units include a first type of grid unit and a second type of grid unit, determining a historical grid output coefficient of each grid unit in the first type of grid unit based on all historical output coefficients of each grid unit in the first type of grid unit, wherein the first type of grid unit is a grid unit having at least one photovoltaic power station distributed therein, and the second type of grid unit is a grid unit having no photovoltaic power station distributed therein;
[0113] Obtain the distribution density of photovoltaic power stations in the target area;
[0114] Based on the geographic information and distribution density, each grid cell in the second type of grid cells is grid mapped to obtain the historical grid output coefficient of each grid cell in the second type of grid cells.
[0115] Optionally, the prediction module 402 may also be specifically used for:
[0116] If the distribution density is greater than a preset value, determining a target grid cell corresponding to each grid cell in the second type of grid cells based on the geographic information, wherein the target grid cell is at least one adjacent grid cell;
[0117] The historical output coefficient or the historical grid output coefficient in the target grid unit is interpolated to obtain the historical grid output coefficient of each grid unit in the second type of grid units.
[0118] Optionally, the prediction module 402 may also be specifically used for:
[0119] If the distribution density is less than or equal to a preset value, obtaining terrain data corresponding to a first specific grid cell among the plurality of grid cells;
[0120] Obtaining spatial distribution characteristics from terrain data corresponding to the first specific grid cell and a specific output coefficient of a second specific grid cell in the first type of grid cells through a preset deep learning model, wherein the specific output coefficient is a historical output coefficient or a historical grid output coefficient;
[0121] The historical grid output coefficient of each grid cell in the second type of grid cells is determined based on the spatial distribution characteristics.
[0122] Optionally, the prediction module 402 may also be specifically used for:
[0123] The historical photovoltaic output coefficient image is analyzed through the preset target output coefficient prediction model to obtain output space information, where the output space information is used to indicate the change of the output coefficient in the time series and the correlation in the spatial distribution;
[0124] Through the target output coefficient prediction model, prediction is made based on the output space information and prediction time to obtain the predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit.
[0125] Optionally, the prediction module 402 may also be specifically used for:
[0126] Performing data preprocessing on the historical power data of each photovoltaic power station to obtain the preprocessed historical power data of each photovoltaic power station;
[0127] The power data at each target time in the preprocessed historical power data of each photovoltaic power station are sorted respectively to obtain the sorted power data at each target time of each photovoltaic power station, wherein the target time is used to indicate the same time of each day in the historical period;
[0128] Determine the theoretical maximum power of each photovoltaic power station at each target time based on the sorted power data of each photovoltaic power station at each target time;
[0129] Based on the actual power value and theoretical maximum power of each PV power station at each target time, the historical output coefficient of each PV power station is calculated.
[0130] Optionally, the prediction module 402 may also be specifically used for:
[0131] Determine the spatial range and resolution of the grid based on the effective power station information, wherein the effective power station information is used to indicate the spatial distribution information and required analysis accuracy of a preset number of photovoltaic power stations that can be determined within a preset distance range;
[0132] Based on the spatial range and resolution, a grid corresponding to the target area is created, wherein the grid includes a plurality of grid cells of uniform size, and each grid cell corresponds to an area in the geographic space;
[0133] Based on the distribution information of photovoltaic power stations at the edge of the target area, the boundary of the grid is adjusted to obtain multiple grid cells corresponding to the target area.
[0134] The functional implementation of each module and each unit in the above-mentioned regional photovoltaic power generation prediction device corresponds to each step in the above-mentioned regional photovoltaic power generation prediction method embodiment, and their functions and implementation processes are no longer repeated here.
[0135] This embodiment also provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the above-mentioned regional photovoltaic power generation power prediction method. The electronic device can be a server or a terminal device.
[0136] See also Figure 5 As shown, the electronic device includes a processor 500 and a memory 501 , wherein the memory 501 stores machine executable instructions that can be executed by the processor 500 , and the processor 500 executes the machine executable instructions to implement the above-mentioned regional photovoltaic power generation power prediction method.
[0137] Further, Figure 5 The electronic device shown further includes a bus 502 and a communication interface 503 , and the processor 500 , the communication interface 503 and the memory 501 are connected via the bus 502 .
[0138] The memory 501 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 503 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 502 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0139] The processor 500 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 500. The above processor 500 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 501, and the processor 500 reads the information in the memory 501 and completes the steps of the regional photovoltaic power generation power prediction method in combination with its hardware.
[0140] The present disclosure also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program runs on a computer, the computer executes the steps of the regional photovoltaic power generation power prediction method.
[0141] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several computer programs to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.
[0143] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure.
Claims
1. A method for predicting regional photovoltaic power generation, characterized in that: The regional photovoltaic power generation power prediction method comprises: Obtain geographic information and historical power data of each photovoltaic power station in the target area; Performing grid output coefficient prediction based on the geographic information and the historical power data to obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate the photovoltaic power generation performance at the prediction time; The predicted power generation power of the target area is calculated based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
2. The method for predicting regional photovoltaic power generation according to claim 1, characterized in that: The calculating the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit includes: Obtaining power generation calculation parameters of each photovoltaic power station in the grid unit, wherein the power generation calculation parameters are installed capacity and / or theoretical maximum power; Calculate the predicted power generation of each photovoltaic power station based on the predicted photovoltaic output coefficient and power generation calculation parameters of each photovoltaic power station in the grid unit; The predicted power generation of all photovoltaic power stations in the target area is integrated to obtain the predicted power generation of the target area.
3. The method for predicting regional photovoltaic power generation according to claim 1, characterized in that: The grid output coefficient prediction based on the geographic information and the historical power data to obtain the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit includes: Determine a historical output coefficient of each photovoltaic power station based on the historical power data of each photovoltaic power station; wherein the historical output coefficient is used to indicate the photovoltaic power generation performance at each historical moment in the historical period; Dividing the target area into grids to obtain a plurality of grid units corresponding to the target area; Based on the geographic information and historical output coefficient of each photovoltaic power station and the plurality of grid cells, a historical photovoltaic output coefficient image is generated; The predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit is obtained by performing prediction based on the historical photovoltaic output coefficient image through a preset target output coefficient prediction model.
4. The method for predicting regional photovoltaic power generation according to claim 3, characterized in that: The generating of a historical photovoltaic output coefficient image based on the geographic information and historical output coefficient of each photovoltaic power station and the plurality of grid units includes: Determining a photovoltaic power station distributed in each grid unit of the plurality of grid units based on geographic information of each photovoltaic power station; Determine the historical grid output coefficient of each grid unit according to the historical output coefficient of the photovoltaic power station distributed in each grid unit; A historical photovoltaic output coefficient image is generated based on the plurality of grid units and the historical grid output coefficient of each grid unit.
5. The method for predicting regional photovoltaic power generation according to claim 4, characterized in that: The determining of the historical grid output coefficient of each grid unit according to the historical output coefficient of the photovoltaic power station distributed in each grid unit includes: If the plurality of grid units include a first type of grid unit and a second type of grid unit, determining a historical grid output coefficient of each grid unit in the first type of grid unit based on all historical output coefficients of each grid unit in the first type of grid unit, wherein the first type of grid unit is a grid unit having at least one photovoltaic power station distributed therein, and the second type of grid unit is a grid unit having no photovoltaic power station distributed therein; Obtaining the distribution density of photovoltaic power stations in the target area; Based on the geographic information and the distribution density, each grid cell in the second type of grid cells is grid mapped to obtain a historical grid output coefficient of each grid cell in the second type of grid cells.
6. The method for predicting regional photovoltaic power generation according to claim 5, characterized in that: The grid mapping is performed on each grid cell in the second type of grid cells based on the geographic information and the distribution density to obtain the historical grid output coefficient of each grid cell in the second type of grid cells, including: If the distribution density is greater than a preset value, determining a target grid cell corresponding to each grid cell in the second type of grid cells based on geographic information, wherein the target grid cell is at least one adjacent grid cell; Interpolation processing is performed on the historical output coefficient or the historical grid output coefficient in the target grid unit to obtain the historical grid output coefficient of each grid unit in the second type of grid units.
7. The method for predicting regional photovoltaic power generation according to claim 5, characterized in that: The grid mapping is performed on each grid cell in the second type of grid cells based on the geographic information and the distribution density to obtain the historical grid output coefficient of each grid cell in the second type of grid cells, including: If the distribution density is less than or equal to a preset value, obtaining terrain data corresponding to a first specific grid cell among the plurality of grid cells; Obtaining spatial distribution characteristics from terrain data corresponding to a first specific grid unit and a specific output coefficient of a second specific grid unit in the first type of grid unit through a preset deep learning model, wherein the specific output coefficient is a historical output coefficient or a historical grid output coefficient; A historical grid output coefficient of each grid cell in the second type of grid cells is determined based on the spatial distribution characteristics.
8. The method for predicting regional photovoltaic power generation according to claim 3, characterized in that: The method of performing prediction based on the historical photovoltaic output coefficient image by using a preset target output coefficient prediction model to obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit includes: By using a preset target output coefficient prediction model, the historical photovoltaic output coefficient image is analyzed to obtain output space information, wherein the output space information is used to indicate the change of the output coefficient in time series and the correlation in spatial distribution; The target output coefficient prediction model is used to perform prediction based on the output space information and the prediction time, so as to obtain the predicted photovoltaic output coefficient corresponding to the prediction time of each photovoltaic power station in the grid unit.
9. The method for predicting regional photovoltaic power generation according to claim 3, characterized in that: Determining the historical output coefficient of each photovoltaic power station based on the historical power data of each photovoltaic power station includes: Performing data preprocessing on the historical power data of each photovoltaic power station to obtain the preprocessed historical power data of each photovoltaic power station; The power data at each target time in the preprocessed historical power data of each photovoltaic power station are sorted respectively to obtain the sorted power data at each target time of each photovoltaic power station, wherein the target time is used to indicate the same time of each day in the historical period; Determine the theoretical maximum power of each photovoltaic power station at each target time based on the sorted power data of each photovoltaic power station at each target time; Based on the actual power value and theoretical maximum power of each PV power station at each target time, the historical output coefficient of each PV power station is calculated.
10. The method for predicting regional photovoltaic power generation according to claim 3, characterized in that: The gridding of the target area to obtain a plurality of grid units corresponding to the target area includes: Determine the spatial range and resolution of the grid based on the effective power station information, wherein the effective power station information is used to indicate the spatial distribution information and required analysis accuracy of a preset number of photovoltaic power stations that can be determined within a preset distance range; Based on the spatial range and resolution, create a grid corresponding to the target area, wherein the grid includes a plurality of grid cells of uniform size, and each grid cell corresponds to an area in the geographic space; Based on the distribution information of the photovoltaic power station at the edge of the target area, the boundary of the grid is adjusted to obtain a plurality of grid units corresponding to the target area.
11. A regional photovoltaic power generation power prediction device, characterized in that: The regional photovoltaic power generation power prediction device comprises: An acquisition module is used to obtain geographic information and historical power data of each photovoltaic power station in the target area; A prediction module, configured to perform grid output coefficient prediction based on the geographic information and the historical power data, and obtain a predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit, wherein the predicted photovoltaic output coefficient is used to indicate the photovoltaic power generation performance at the prediction moment; The calculation module is used to calculate the predicted power generation power of the target area based on the predicted photovoltaic output coefficient of each photovoltaic power station in the grid unit.
12. An electronic device, characterized in that: The electronic device includes a processor and a memory, the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the regional photovoltaic power generation power prediction method described in any one of claims 1-10.
13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting regional photovoltaic power generation as described in any one of claims 1 to 10 is implemented.
Citation Information
Cited By
Photovoltaic power prediction method and discharge control method of energy storage system and energy storage system
CN119726713A
