Power prediction method and system for distributed photovoltaic power station

Abstract

A power prediction method and system for a distributed photovoltaic power station, relating to the technical field of photovoltaic power station power prediction. The power prediction method comprises determining a historical data set of a distributed photovoltaic power station in a first preset time period based on a start time; determining a future irradiance data set of a distributed photovoltaic power station in a second preset time period based on a start time; determining a first future power generation data set of a second preset time period based on a first preset algorithm according to the historical data set and the future irradiance data set; determining a second future power generation data set of a second preset time period based on a second preset algorithm according to the historical data set and the future irradiance data set; determining a comprehensive future power generation data set of a distributed photovoltaic power station based on the first future power generation data set and the second future power generation data set. The prediction method can improve the accuracy of prediction, reduce the uncertainty of photovoltaic power generation, and improve the ability of the power grid to accept randomly fluctuating power sources.

Description

Technical Field

[0001] The present application relates to the technical field of photovoltaic power station power prediction, and in particular to a power prediction method and system for a distributed photovoltaic power station. Background Art

[0002] Distributed photovoltaic power stations are affected by factors such as irradiance and photovoltaic module temperature, which causes their output power to have obvious volatility and intermittent characteristics. Predicting their power can help balance electricity supply and demand and avoid mismatches in electricity supply and demand.

[0003] The existing power prediction method of distributed photovoltaic power stations generally calculates the predicted power generation by predicting irradiance. However, the inventors of this application have found that due to the wide distribution of distributed photovoltaic power stations, the accuracy of predicting power generation by only predicting irradiance is not high.

[0004] The contents of the background technology are merely technologies known to the public and do not necessarily represent the existing technologies in the field. Summary of the invention

[0005] The present application aims to provide a power prediction method and system for a distributed photovoltaic power station to solve the above-mentioned technical problems.

[0006] According to one aspect of the present application, the present application provides a power prediction method for a distributed photovoltaic power station. The power prediction method includes: based on the start time, determining a historical data set of the distributed photovoltaic power station in a first preset time period, the historical data set including historical predicted irradiance data determined according to the first preset time period, actual power generation data, and first moment data corresponding to the historical predicted irradiance data and actual power generation data; based on the start time, determining a future irradiance data set of the distributed photovoltaic power station in a second preset time period, the future irradiance data set including future predicted irradiance data determined according to the second preset time period, and second moment data corresponding to the future predicted irradiance data; based on the historical data set and the future irradiance data set, determining a first future power generation data set for the second preset time period based on a first preset algorithm, the first future power generation data set including the second moment data corresponding to each second preset time period. The invention relates to a method for determining a second future power generation data set for a second preset time period based on a historical data set and a future irradiance data set, and a second moment data corresponding to the first future power generation data set; determining a second future power generation data set for a second preset time period based on a second preset algorithm according to the historical data set and the future irradiance data set, the second future power generation data set including the second future power generation data corresponding to each second preset time period, and the second moment data corresponding to the second future power generation data; determining a comprehensive future power generation data set of a distributed photovoltaic power station according to the first future power generation data set and the second future power generation data set, the comprehensive future power generation data set including the comprehensive future predicted power generation data corresponding to each second preset time period, and the second moment data corresponding to the comprehensive future predicted power generation data.

[0007] According to some embodiments of the present application, the step of determining a historical data set of a distributed photovoltaic power station in a first preset time period based on a start time includes: determining an initial historical data set of the distributed photovoltaic power station in the first preset time period based on the start time; and performing data processing on the initial historical data set according to a first preset processing rule to obtain a historical data set.

[0008] According to some embodiments of the present application, the step of determining a first future power generation data set for a second preset time period based on a first preset algorithm according to a historical data set and a future irradiance data set includes: determining a first prediction data table according to the historical data set; performing data processing on the first prediction data table according to a second preset processing rule to obtain a first component of the first future power generation data set; performing data processing on the historical data set and the first prediction data table according to a third preset processing rule to obtain a second component of the first future power generation data set; and determining the first future power generation data set based on the first component and the second component.

[0009] According to some embodiments of the present application, after the step of determining a first future power generation data set based on the first component and the second component, the power prediction method further includes: when the historical data set and the first prediction data table meet the first preset extreme weather condition, performing data processing on the historical data set and the first prediction data table according to a fourth preset processing rule to update the first future power generation data set.

[0010] According to some embodiments of the present application, after the step of determining a first future power generation data set based on the first component and the second component, the power prediction method also includes: when the historical data set and the first prediction data table meet the second preset extreme weather condition, performing data processing on the historical data set and the first prediction data table according to a fifth preset processing rule to update the first future power generation data set.

[0011] According to some embodiments of the present application, the step of determining a second future power generation data set for a second preset time period based on a second preset algorithm according to a historical data set and a future irradiance data set includes: determining a second prediction data table and a third prediction data table according to the historical data set; and performing data processing on the second prediction data table and the third prediction data table according to a sixth preset processing rule to obtain a second future power generation data set.

[0012] According to some embodiments of the present application, after the step of performing data processing on the second prediction data table and the third prediction data table according to the sixth preset processing rule to obtain a second future power generation power data set, the power control method also includes: when the second prediction data table and the third prediction data table meet the first preset correction condition, performing data processing on the second prediction data table and the third prediction data table according to the seventh preset processing rule to obtain a fitting value of the second future power generation power data set; updating the second future power generation power data set based on the fitting value of the second future power generation power data set and the second future power generation power data set.

[0013] According to some embodiments of the present application, after the step of updating the second future power generation power data set based on the fitting value of the second future power generation power data set and the second future power generation power data set, the power prediction method also includes: when the second prediction data table and the third prediction data table meet the second preset correction condition, performing data processing on the second prediction data table and the third prediction data table according to an eighth preset processing rule to update the second future power generation power data set.

[0014] According to some embodiments of the present application, the step of determining a comprehensive future power generation data set of a distributed photovoltaic power station based on a first future power generation data set and a second future power generation data set includes: determining a comprehensive future power generation data set based on the first future power generation data set and corresponding coefficients, the second future power generation data set and corresponding coefficients.

[0015] According to another aspect of the present application, the present application provides a power prediction system for a distributed photovoltaic power station. The power prediction system is used to execute the power prediction method for a distributed photovoltaic power station as described above, and the power prediction system includes a processing unit. The processing unit determines a historical data set of the distributed photovoltaic power station in a first preset time period based on a start time, the historical data set includes historical predicted irradiance data determined according to a first preset time period, actual power generation data, and first moment data corresponding to the historical predicted irradiance data and actual power generation data; the processing unit determines a future irradiance data set of the distributed photovoltaic power station in a second preset time period based on a start time, the future irradiance data set includes future predicted irradiance data determined according to a second preset time period, and second moment data corresponding to the future predicted irradiance data; the processing unit determines a first future power generation data set for a second preset time period based on a first preset algorithm according to the historical data set and the future irradiance data set, the first future power generation data set includes data corresponding to each second preset time period. The processing unit determines a second future power generation data set for a second preset time period based on a second preset algorithm according to the historical data set and the future irradiance data set, the second future power generation data set including the second future power generation data corresponding to each second preset time period, and the second moment data corresponding to the second future power generation data; the processing unit determines a comprehensive future power generation data set of the distributed photovoltaic power station according to the first future power generation data set and the second future power generation data set, the comprehensive future power generation data set including the comprehensive predicted power generation data corresponding to each second preset time period, and the second moment data corresponding to the comprehensive predicted power generation data.

[0016] According to another aspect of the present application, the present application further provides a non-volatile computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the power prediction method of the distributed photovoltaic power station as described above can be implemented.

[0017] According to another aspect of the present application, the present application also provides an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the one or more processors are enabled to implement the power prediction method of the distributed photovoltaic power station as described above.

[0018] According to another aspect of the present application, the present application also provides a computer program product, including: a computer program stored on a computer-readable storage medium; the computer program includes program instructions, and when the program instructions are executed by the computer, the computer executes the power prediction method of the distributed photovoltaic power station as described above.

[0019] Beneficial Effects

[0020] The power prediction method provided in the present application can determine a first future power generation data set according to a historical data set and a future irradiance data set through a first preset algorithm. The present application can determine a second future power generation data set according to a historical data set and a future irradiance data set through a second preset algorithm. The present application can determine a comprehensive future power generation data set according to the first future power generation data set and the second future power generation data set.

[0021] The power prediction method provided in this application can predict the power generation of distributed photovoltaic power stations through historical predicted irradiance data, actual power generation data and future irradiance data sets. This application can predict the power generation of distributed photovoltaic power stations through two algorithms, which can improve the accuracy of prediction, reduce the uncertainty of photovoltaic power generation, and improve the ability of the power grid to accept random fluctuating power sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A schematic diagram showing a flow chart of a power prediction method 1000 according to an embodiment of the present application is shown;

[0024] Figure 2 A schematic diagram showing the process of step S110 according to an embodiment of the present application is shown;

[0025] Figure 3 A schematic diagram showing the process of step S130 according to an embodiment of the present application is shown;

[0026] Figure 4Another schematic diagram of the process of step S130 according to an embodiment of the present application is shown;

[0027] Figure 5 A schematic diagram showing the process of step S140 according to an embodiment of the present application is shown;

[0028] Figure 6 Another schematic diagram of the process of step S140 according to an embodiment of the present application is shown;

[0029] Figure 7 A schematic diagram showing the process of step S150 according to an embodiment of the present application is shown;

[0030] Figure 8 A schematic diagram of the structure of a power prediction system according to an embodiment of the present application is shown.

[0031] Reference numerals:

[0032] Power prediction system 200 ; processing unit 210 . DETAILED DESCRIPTION

[0033] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0034] The described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of these specific details, or other modes, components, materials, devices, etc. may be adopted. In these cases, known structures, methods, devices, implementations, materials or operations will not be shown or described in detail.

[0035] In addition, the terms "include" and "have" 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 limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0036] The terms "first", "second" and the like in the specification and claims of this application and the above drawings are used to distinguish different objects rather than to describe a specific order.

[0037] The following is a clear and complete description of the technical solution of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0038] According to one aspect of the present application, the present application provides a power prediction method 1000 for a distributed photovoltaic power station. The power prediction method 1000 may be executed by a power prediction system, and illustratively, the power prediction system may be a host (or server) with data processing capabilities.

[0039] See also Figure 1 , the power prediction method 1000 may include steps S110 to S150.

[0040] In step S110, the power prediction system determines a historical data set of the distributed photovoltaic power station in the first preset time period based on the start time. The historical data set includes historical predicted irradiance data, actual power generation data, and first time data corresponding to the historical predicted irradiance data and actual power generation data determined according to the first preset time period.

[0041] According to an example embodiment, the start time may be a start time when it is desired to predict the power generation of the distributed photovoltaic power station.

[0042] The first preset time period may be a time period corresponding to the selected historical data set of the distributed photovoltaic power station. The first preset time period may be a time period taken forward from the start time. The forward time period and the time resolution may be determined according to the region where the distributed photovoltaic power station is located and the local climate conditions.

[0043] Exemplarily, the forward time period may be 70 days, 90 days, 100 days, etc.

[0044] According to an example embodiment, the historical data set may be a set of historical operating status data of a distributed photovoltaic power station.

[0045] For example, the first preset time period may be a time period for acquiring historical data, and the first preset time period may be 1 minute, 5 minutes, or 15 minutes.

[0046] The predicted irradiance may be the radiant energy density received per unit area of ​​the distributed photovoltaic power station. The historical predicted irradiance data may be the predicted irradiance data of the distributed photovoltaic power station within a first preset time. The historical predicted irradiance data may be determined by a predicted irradiance algorithm.

[0047] The actual power generation may be the actual output power value of the distributed photovoltaic power station. The actual power generation data may be the actual power generation data of the distributed photovoltaic power station within the first preset time. The actual power generation data may be obtained through the power collection device of the distributed photovoltaic power station.

[0048] The first moment data may be time data corresponding to the historical predicted irradiance data and the actual generated power data.

[0049] For example, the first preset time period can be 70 days, and the first preset time period can be 15 minutes. The power prediction system can round the start time as the first moment (the first moment is recorded as TNow), take 15 minutes as a node, and take the historical predicted irradiance data, actual power generation data, and first moment data of the distributed photovoltaic power station for 70 days from the first moment forward to form a historical data set.

[0050] In step S120, the power prediction system determines a future irradiance data set of the distributed photovoltaic power station in a second preset time period based on the start time. The future irradiance data set includes future predicted irradiance data determined according to the second preset time period and second time data corresponding to the future predicted irradiance data.

[0051] The second preset time period may be a time period for predicting the power generation of the distributed photovoltaic power station. The second preset time period may include an ultra-short-term prediction time period and a short-term prediction time period. The ultra-short-term prediction time period may generally be several hours. The short-term prediction time period may be several days.

[0052] The future irradiance dataset can be a collection of future predicted irradiance data for distributed photovoltaic power stations.

[0053] The second preset time period may be a time period for predicting the power generation of the distributed photovoltaic power station. The second preset time period may be 1 minute, 5 minutes, or 15 minutes, etc. The second preset time period may correspond to the first preset time period.

[0054] The future predicted irradiance data may be the predicted irradiance data of the distributed photovoltaic power station within the second preset time. The future predicted irradiance data may be determined by a predicted irradiance algorithm.

[0055] The second moment data may be time data corresponding to the future predicted irradiance data.

[0056] For example, the ultra-short-term prediction time period can be 4 decimals, and the second preset time period can be 15 minutes. The power prediction system can round the start time as the first moment, take 15 minutes as a node, and take the future predicted irradiance data of the distributed photovoltaic power station for 4 hours (i.e. 16 moments) from the first time and the second moment data to form a future irradiance data set.

[0057] For another example, the short-term prediction period can be 7 days, and the second preset time period can be 15 minutes. The power prediction system can round the start time as the first moment, take 15 minutes as a node, and take the future predicted irradiance data of the distributed photovoltaic power station for 7 days (i.e. 16 moments) from the first time and the second moment data to form a future irradiance data set.

[0058] In step S130, the power prediction system determines a first future power generation data set for a second preset time period based on a first preset algorithm according to the historical data set and the future irradiance data set. The first future power generation data set includes first future power generation data corresponding to each second preset time period, and second moment data corresponding to the first future power generation data.

[0059] According to an example embodiment, the first preset algorithm may be a mean value algorithm. The first future power generation data set may be a predicted power generation data set of a distributed photovoltaic power station within a second preset time period calculated by the first preset algorithm.

[0060] The first future power generation data may be the predicted power generation data of the distributed photovoltaic power station within the second preset time period calculated by the first preset algorithm. The first future power generation data corresponds to the second moment data.

[0061] For example, the power prediction system can calculate the first future power generation data of the distributed photovoltaic power station within the ultra-short-term prediction time period through a first preset algorithm. The power prediction system can calculate the first future power generation data of the distributed photovoltaic power station within the short-term prediction time period through a first preset algorithm.

[0062] In step S140, the power prediction system determines a second future power generation data set for a second preset time period based on a second preset algorithm according to the historical data set and the future irradiance data set. The second future power generation data set includes second future power generation data corresponding to each second preset time period, and second moment data corresponding to the second future power generation data.

[0063] According to an example embodiment, the second preset algorithm may be a regression algorithm. The second future power generation data set may be a predicted power generation data set of a distributed photovoltaic power station within a second preset time period calculated by the second preset algorithm.

[0064] The second future power generation data may be the predicted power generation data of the distributed photovoltaic power station within the second preset time period calculated by the second preset algorithm. The second future power generation data corresponds to the second moment data.

[0065] For example, the power prediction system can calculate the second future power generation data of the distributed photovoltaic power station within the ultra-short-term prediction time period by a second preset algorithm. The power prediction system can calculate the second future power generation data of the distributed photovoltaic power station within the short-term prediction time period by a second preset algorithm.

[0066] In step S150, the power prediction system determines a comprehensive future power generation data set of the distributed photovoltaic power station based on the first future power generation data set and the second future power generation data set. The comprehensive future power generation data set includes comprehensive future predicted power generation data corresponding to each second preset time period, and second time data corresponding to the comprehensive future predicted power generation data.

[0067] According to an example embodiment, the comprehensive future power generation data set may be a set of predicted power generation data of a distributed photovoltaic power station within a second preset time period determined according to the first future power generation data set and the second future power generation data set.

[0068] The comprehensive future power generation data may be predicted power generation data of the distributed photovoltaic power station within a second preset time period determined according to the first future power generation data set and the second future power generation data set.

[0069] For example, the power prediction system may determine a comprehensive future power generation data set based on the first future power generation data set and the corresponding coefficients, the second future power generation data set and the corresponding coefficients.

[0070] Through the above embodiments, the power prediction method 1000 provided in the present application can determine the first future power generation data set according to the historical data set and the future irradiance data set through the first preset algorithm. The present application can determine the second future power generation data set according to the historical data set and the future irradiance data set through the second preset algorithm. The present application can determine the comprehensive future power generation data set according to the first future power generation data set and the second future power generation data set.

[0071] The power prediction method 1000 provided in this application can predict the power generation of distributed photovoltaic power stations through historical predicted irradiance data, actual power generation data and future irradiance data sets. This application can predict the power generation of distributed photovoltaic power stations through two algorithms, which can improve the accuracy of prediction, reduce the uncertainty of photovoltaic power generation, and improve the ability of the power grid to accept random fluctuating power sources.

[0072] Alternatively, see Figure 2 , step S110 may include step S111 and step S112.

[0073] In step S111 , the power prediction system determines an initial historical data set of the distributed photovoltaic power station in a first preset time period based on the start time.

[0074] According to an example embodiment, the initial historical data set is an initial historical operating status data set of a distributed photovoltaic power station. The initial historical data set may include historical predicted irradiance data, actual power generation data, and first time point data corresponding to the historical predicted irradiance data and actual power generation data determined according to a first preset time period.

[0075] In step S112, the power prediction system processes the initial historical data set according to a first preset processing rule to obtain a historical data set.

[0076] According to an example embodiment, the first preset processing rule may be a preset rule for performing data processing on the initial historical data set.

[0077] For example, the first preset rule may include:

[0078] 1. Delete the entire row of data corresponding to the empty data in the initial historical data set.

[0079] 2. Set the negative value of actual power generation to 0.

[0080] 3. Arrange the initial historical data set in reverse chronological order.

[0081] 4. Determine the historical predicted radiation illumination data and actual power generation data at the corresponding moment in the first specific time period.

[0082] 5. Take integers for the historical predicted radiation illumination data and the actual power generation data, and process the historical predicted radiation illumination data and the actual power generation data within the second specific time period according to the first preset deletion rule.

[0083] The power prediction system generates a historical data set after processing the initial historical data set according to the first preset processing rule.

[0084] According to an example embodiment, the first specific time period may be a time interval determined according to a time period during which the distributed photovoltaic power station can generate electricity. For example, the first specific time period may be 18 hours.

[0085] The start time of the first specific time period may be determined according to the region where the distributed photovoltaic power station is located. For example, the start time of the first specific time period may be 4:30, and the power prediction system determines the first specific time period as [4:30, 22:30].

[0086] The power prediction system screens the first moment data in the initial historical data set, and the power prediction system only retains the historical predicted radiation illumination data and actual power generation data corresponding to the first moment data within the first specific time period.

[0087] The power prediction system records the number of data points of the historical predicted irradiance data and the number of data points of the actual power generation data in the first specific period as NN. (NN = the number of data points of the historical predicted irradiance data in the first specific period = the number of data points of the actual power generation data in the first specific period).

[0088] For example, when the first specific period is 18 hours, NN is 73.

[0089] According to an example embodiment, the rounding may be rounding down to an integer. The power prediction system rounds down the value of the historical predicted irradiance data. The power prediction system rounds down the value of the actual power generation data.

[0090] For example, the power prediction system may retain only integer values ​​of the predicted irradiance value, and the power prediction system may retain only integer values ​​of the actual generated power value.

[0091] According to an example embodiment, the second specific time period may be a time interval determined according to a time period when the distributed photovoltaic power station has a higher power generation efficiency. For example, the second specific time period may be 7.5 hours.

[0092] The start time of the second specific time period may be determined according to the region where the distributed photovoltaic power station is located. For example, the start time of the second specific time period may be 9:00, and the power prediction system determines the second specific time period as [9:00, 15:30].

[0093] The first preset deletion rule may be a preset rule for processing the initial historical data set after filtering and rounding. For example, the first preset deletion rule may be: when the value of the actual power generation data is 0 and the value of the actual power generation data is a repeated value, delete the entire row of data corresponding to the actual power generation data.

[0094] The power forecasting system screens the first moment data in the initial historical data set, and the power forecasting system determines the actual power generation data of the first moment data in the second specific time period. When the power forecasting system determines that the value of the actual power generation data is 0, the power forecasting system deletes the entire row of data corresponding to the value of the actual power generation data being 0. When the power forecasting system determines that the value of the actual power generation data is a repeated value, the power forecasting system deletes the entire row of data corresponding to the value of the actual power generation data being 0.

[0095] When the power prediction system determines that the number of repetitions of the actual power generation data is greater than or equal to 4, the power prediction system determines that the actual power generation data is a repeated value.

[0096] Optionally, in step S110, the power prediction system may record the historical predicted irradiance data in the historical data set as Ra0, and the actual generated power data as Rp0.

[0097] Optionally, in step S120, the power prediction system may record the future predicted irradiance data within the ultra-short-term prediction period obtained as R H . The power prediction system can also retrieve the all-day future predicted irradiance data set corresponding to the 16th moment. The all-day future predicted irradiance data set includes the all-day future predicted irradiance data corresponding to each second preset period, and the third moment data corresponding to the all-day future predicted irradiance data. The all-day future predicted irradiance data set can include 96 all-day future predicted irradiance data and 96 third moment data. The all-day future predicted irradiance data is recorded as NWP.

[0098] The power prediction system may obtain 96×7 future predicted irradiance data within the short-term prediction period, and the future predicted irradiance data is recorded as NWP_DQ.

[0099] Alternatively, see Figure 3 , step S130 may include steps S131 to S134.

[0100] In step S131 , the power prediction system determines a first prediction data table according to the historical data set.

[0101] According to an example embodiment, the first prediction data table may be a prediction data table determined according to a historical data set and a first preset coefficient. For example, the first preset coefficient may be set to 8, and the power prediction system may take the first NN×8 rows of data in the historical data set to form the first prediction data table.

[0102] In step S132, the power prediction system performs data processing on the first prediction data table according to a second preset processing rule to obtain a first component of a first future generated power data set.

[0103] The second preset processing rule may be a preset rule for performing data processing on the historical data set and the first prediction data table.

[0104] For example, for an ultra-short-term prediction period, the second preset processing rule may include:

[0105] 1. The power prediction system sums up the historical predicted irradiance data in the historical data set by day and finds the date with the maximum 5 days.

[0106] 2. The power prediction system sums up the actual power generation data in the historical data set on a daily basis to find the date with the maximum of 5 days; and takes the intersection of the above two sets of dates.

[0107] 3. After the power forecasting system retrieves the intersection date, it retrieves the historical forecast irradiance data (Ra1) and actual power generation data (Rp1) of the corresponding date in the first forecasting data table according to the intersection date;

[0108] 4. The power prediction system uses Ra1 and the first-order polynomial least squares regression to fit Rp1 (Y=aX+b, Ra1 is substituted into X, Rp1 is substituted into Y, and the optimization coefficients of a and b are obtained).

[0109] 5. The future predicted irradiance data (denoted as R H ) is substituted into X for calculation, and the first components (denoted as P1) of 16 corresponding first future power generation data are obtained.

[0110] The 16 corresponding first components of the first future power generation data and the second moment data corresponding to the first components of the first future power generation data form the first component of the first future power generation data set in the ultra-short-term prediction time period.

[0111] For the short-term forecast period, the above R H By replacing it with NWP_DQ, the first component (also denoted as P1) of the first future power generation data of the short-term prediction period can be obtained. The power prediction system forms the first component of the first future power generation data set of the short-term prediction period according to the first component of the first future power generation data and the second moment data corresponding to the first component of the first future power generation data.

[0112] In step S133, the power prediction system performs data processing on the historical data set and the first prediction data table according to the third preset processing rule to obtain the second component of the first future generated power data set.

[0113] The third preset processing rule may be a preset rule for performing data processing on the historical data set and the first prediction data table.

[0114] For example, the third preset processing rule may include:

[0115] 1. The historical forecast irradiance data in the first forecast data table are summed up on a daily basis, and the power forecast system retrieves the date of the maximum 6 days after the historical forecast irradiance data is summed up on a daily basis.

[0116] 2. Calculate the average value of the actual power generation data of these 6 days according to the data at the second moment, and obtain the mean value of the actual power generation data of each second moment (denoted as rpmean, the number of rpmeans should be NN).

[0117] 3. Add 18 zeros before rpmean and 5 zeros after it to form 96 actual power generation data points for one day (recorded as rpmean1). The 96 actual power generation data points all correspond to the time data one by one.

[0118] For the ultra-short-term prediction period, the power prediction system extracts the actual power generation data corresponding to the second moment data from rpmean1 to obtain 16 corresponding second components of the first future power generation data (denoted as P2).

[0119] The power prediction system can form the second component of the first future power generation data set of the ultra-short-term prediction time period based on the second components of the 16 corresponding first future power generation data and the second moment data corresponding to the first future power generation data.

[0120] For the short-term prediction period, the rpmean1 is used to replace the first component of the first future power generation data of the short-term prediction period (P1) of the first future power generation data of the day after the start time, thereby obtaining the second component of the first future power generation data (also recorded as P2). The power prediction system can form the second component of the first future power generation data set of the short-term prediction period based on the second component of the first future power generation data and the second moment data corresponding to the first future power generation data.

[0121] In step S134, the power prediction system determines a first future power generation data set according to the first component and the second component.

[0122] According to an example embodiment, a power prediction system determines a first future generation power data set based on a first component and a corresponding coefficient, and a second component and a corresponding coefficient.

[0123] Due to the relatively scattered distribution of distributed photovoltaic power stations, the result determined by historical predicted irradiance data accounts for a relatively small proportion, that is, the proportion of the first component of the first future power generation power dataset is relatively small. The corresponding coefficient of the first component of the first future power generation power dataset and the corresponding coefficient of the second component of the first future power generation power dataset can be adjusted according to the natural environment where the distributed photovoltaic power station is located. For example, the corresponding coefficient of the first component of the first future power generation power dataset can be taken as 0.2, and the corresponding coefficient of the second component of the first future power generation power dataset can be taken as 0.8.

[0124] The power prediction system can determine the first future power generation data according to the following formula:

[0125] UST_F1 = 0.2×p1 + 0.8×p2;

[0126] Where UST_F1 is the first future power generation data; p1 is the first component of the first future power generation data; p2 is the second component of the first future power generation data.

[0127] The power prediction system can form the first future power generation power dataset according to the first future power generation data and the second moment data corresponding to the first future power generation data.

[0128] Optionally, referring to Figure 4 , step S130 may further include step S135.

[0129] In step S135, when the historical dataset and the first prediction data table meet the first preset extreme weather condition, the power prediction system processes the historical dataset and the first prediction data table according to the fourth preset processing rule to update the first future power generation power dataset.

[0130] According to the exemplary embodiment, the first preset extreme weather condition may be a condition for determining that the data in the historical dataset and the first prediction data table are extreme overcast data.

[0131] The power prediction system can take out the maximum historical predicted irradiance data (denoted as MRD) of each day in the first prediction data table and the number of data points of each day. If the number of data points of a certain day is less than NN - 10, then the maximum historical predicted irradiance data of this day is set to 0.

[0132] The power prediction system records all the numbers greater than 0 in MRD as MRD1, and records the corresponding days of MRD1 as RiQi1.

[0133] For the ultra - short - term prediction time period, the first preset extreme weather condition may be: RiQi1 is greater than or equal to the first preset coefficient (8) - 1, and max(NWP) < the average value of the first 2 numerical values of MRD1.

[0134] For a short - term prediction time period, the first preset extreme weather condition can be: RiQi1 is greater than or equal to the first preset coefficient (8) - 1, and the average value of the first 2 numbers of max(NWP_DQ) < MRD1.

[0135] The fourth preset processing rule can be a preset rule for processing data of the historical data set and the first prediction data table.

[0136] When the power prediction system determines that the historical data set and the first prediction data table meet the first preset extreme weather condition, the power prediction system processes the historical data set and the first prediction data table according to the fourth preset processing rule.

[0137] For an ultra - short - term prediction time period, the fourth prediction processing rule can be:

[0138] 1. The power prediction system extracts the historical predicted irradiance data (Ra2) and the actual power generation data (Rp2) corresponding to the first 2 days containing RiQi1 in the first prediction data table.

[0139] 2. If the number of points of the actual power generation data (Rp2) of the first 2 days containing RiQi1 is greater than or equal to NN×2 - 6, the power prediction system uses the Ra2 and performs a least - squares regression fitting of Rp2 with a first - order polynomial (Y = cX + d, substituting Ra2 into X and Rp2 into Y to find the optimization coefficients c and d).

[0140] 3. Substitute the future predicted irradiance data (denoted as R H ) into X for calculation to obtain 16 corresponding first - future power generation data. Among them, when the future predicted irradiance data (R H ) is 0, the first - future power generation data is also 0.

[0141] The 16 corresponding first - future power generation data, and the second - moment data corresponding to the first - future power generation data form the first - future power generation data set for the ultra - short - term prediction time period, which is used to update the first - future power generation data set for the ultra - short - term prediction time period in step S134.

[0142] If the number of points of the actual power generation data (Rp) of the first 2 days containing RiQi1 is less than NN×2 - 6, the power prediction system does not update the first - future power generation data set for the ultra - short - term prediction time period in step S134.

[0143] For the short - term prediction time period, the power prediction system uses the above - mentioned R HBy replacing it with NWP_DQ, the first future power generation data of the short-term prediction period can be obtained, thereby forming the first future power generation data set of the short-term prediction period to update the first future power generation data set of the short-term prediction period in step S134.

[0144] When the power prediction system determines that the historical data set and the first prediction data table do not satisfy the first preset extreme weather condition, the power prediction system does not execute step S135.

[0145] Alternatively, see Figure 4 , step S130 may further include step S136.

[0146] In step S136, when the historical data set and the first prediction data table meet the second preset extreme weather condition, the power prediction system processes the historical data set and the first prediction data table according to the fifth preset processing rule to update the first future power generation data set.

[0147] The second preset extreme weather condition may be a condition for determining whether the data in the historical data set and the data in the first prediction data table are extreme sunny day data.

[0148] For the ultra-short-term prediction period, the second preset extreme weather condition may be: max(NWP) is greater than the average of the first two numbers of MRD.

[0149] The fifth preset processing rule may be a preset rule for performing data processing on the historical data set and the first prediction data table.

[0150] When the power prediction system determines that the historical data set and the first prediction data table meet the second preset extreme weather condition, the power prediction system performs data processing on the historical data set and the first prediction data table according to the fifth preset processing rule.

[0151] The fifth prediction processing rule may be:

[0152] 1. The power forecasting system takes out the actual power generation data (Rp) of the dates corresponding to the top three maximum values ​​of MRD, and calculates the average value of Rp according to the data at the second moment, which is the mean value of the actual power generation data of each second moment data (denoted as rpmean2, the number of rpmean2 should be NN).

[0153] 2. Add 18 zeros before rpmean2 and 5 zeros after it to form 96 actual power generation data points for one day (recorded as rpmean3). The 96 actual power generation data points all correspond to the time data one by one.

[0154] For the ultra-short-term prediction time period, the power prediction system determines that sum(the value of the ultra-short-term prediction time period in rpmean3)>sum(UST_F1 of the ultra-short-term prediction time period in step S134), then the power prediction system takes out the actual power generation data corresponding to the second moment data from rpmean3, and obtains 16 corresponding first future power generation data, among which the future predicted irradiance data (R H ) is 0, the first future power generation data is also 0.

[0155] The 16 corresponding first future power generation data and the second moment data corresponding to the first future power generation data form a first future power generation data set for an ultra-short-term prediction period, which is used to update the first future power generation data set for an ultra-short-term prediction period in step S134.

[0156] For the short-term prediction time period, the power prediction system determines that sum(the value of rpmean3 corresponding to the short-term prediction time period)>sum(UST_F1 of the short-term prediction time period in step S134), then the power prediction system will replace rpmean3 with the first future power generation power data of the day after the start time in UST_F1 of the short-term prediction time period in step S134, thereby updating the first future power generation power data set of the short-term prediction time period in step S134.

[0157] When the power prediction system determines that the historical data set and the first prediction data table do not satisfy the second preset extreme weather condition, the power prediction system does not execute step S136.

[0158] Alternatively, see Figure 5 , step S140 may also include step S141 and step S142.

[0159] In step S141 , the power prediction system determines a second prediction data table and a third prediction data table according to the historical data set.

[0160] According to an example embodiment, the second prediction data table may be a prediction data table determined based on the historical data set and the second preset coefficient. For example, the second preset coefficient may be set to 6, and the power prediction system may take the first NN×6 rows of data in the historical data set to form the second prediction data table.

[0161] The third prediction data table may be a prediction data table determined according to the historical data set and the third preset coefficient. For example, the third preset coefficient may be set to 24, and the power prediction system may take the first NN×24 rows of data in the historical data set to form the third prediction data table.

[0162] In step S142, the power prediction system processes the second prediction data table and the third prediction data table according to the sixth preset processing rule to obtain a second future power generation data set.

[0163] The sixth preset processing rule may be a preset rule for performing data processing on the second prediction data table and the third prediction data table.

[0164] For example, for an ultra-short-term prediction period, the sixth preset processing rule may include:

[0165] 1. The power prediction system divides the sum of all future predicted irradiance data in the third prediction data table by the total number of days in the third prediction data table to obtain the mean future predicted irradiance of the third prediction data table, which is recorded as sumnwp.

[0166] 2. The power prediction system takes sumnwp_now=sum(NWP).

[0167] 3. If the power prediction system determines that sumnwp_now≥sumnwp×regression method cloudy and sunny day distribution ratio (which can be 0.8), the fourth prediction data table is taken as the second prediction data table. If the power prediction system determines that sumnwp_now<sumnwp×regression method cloudy and sunny day distribution ratio (which can be 0.8), the fourth prediction data table is taken as the third prediction data table.

[0168] 4. The power prediction system takes out the historical predicted irradiance data (Ra3) and the actual power generation data (Rp3) in the fourth prediction data table; the power prediction system uses Ra3 and quadratic polynomial least squares regression to fit Rp3 (Y=e*X²+f*X+g, Ra3 is substituted with X, Rp3 is substituted with Y, and the optimization coefficients of e, f and g are obtained).

[0169] 5. The future predicted irradiance data (denoted as R H ) is substituted into X for calculation, and 16 corresponding second future power generation data (denoted as UST_F2) are obtained.

[0170] The power prediction system can form a second future power generation data set of an ultra-short-term prediction period according to the 16 corresponding second future power generation data and the second moment data corresponding to the second future power generation data.

[0171] For the short-term prediction period, the above power prediction system takes sumnwp_now=sum(future predicted irradiance data of the next day after the start time); HBy replacing NWP_DQ, the second future power generation data (also recorded as UST_F2) of the short-term prediction period can be obtained. The power prediction system can form a second future power generation data set of the short-term prediction period according to the second future power generation data and the second moment data corresponding to the second future power generation data.

[0172] Alternatively, see Figure 6 , step S140 may also include step S143-step S144.

[0173] In step S143, when the second prediction data table and the third prediction data table meet the first preset correction condition, the power prediction system processes the second prediction data table and the third prediction data table according to the seventh preset processing rule to obtain the fitting value of the second future power generation data set.

[0174] According to an example embodiment, the first preset correction condition may be a condition for performing symmetry axis offset correction on the second future power generation data set.

[0175] The power prediction system substitutes the historical predicted irradiance data (Ra3) in the fourth prediction data table into X in Y=e*X²+f*X+g to obtain a set of power generation data (denoted as fy). The power prediction system takes the maximum value of fy and takes the row mark of the maximum value of fy in fy (denoted as local). The power prediction system takes the actual power generation data of the local row of the fourth prediction data table (denoted as maxx).

[0176] The power prediction system takes the maximum historical predicted irradiance data (denoted as maxr) in the fourth prediction data table.

[0177] For example, the first preset correction condition may be: maxr>(maxx+regression method symmetry axis offset), and the regression method symmetry axis offset may be 150.

[0178] When the power prediction system determines that the second prediction data table and the third prediction data table meet the first preset correction condition, the power prediction system performs data processing on the second prediction data table and the third prediction data table according to the seventh preset processing rule.

[0179] The seventh preset processing rule may be a preset rule for performing data processing on the second prediction data table and the third prediction data table.

[0180] For example, for an ultra-short-term prediction period, the seventh preset processing rule may be:

[0181] 1. The power prediction system retrieves the historical prediction irradiance data (Ra3) and the actual power generation data (Rp3) from the fourth prediction data table.

[0182] 2. The power prediction system uses Ra3 and the first-order polynomial least squares regression to fit Rp3 (Y=hX+i, Ra3 is substituted with X, Rp3 is substituted with Y, and the optimization coefficients of h and i are obtained).

[0183] 3. The future predicted irradiance data (denoted as R H ) is substituted into X for calculation, and 16 corresponding fitting values ​​of the second future power generation data (denoted as f1) are obtained.

[0184] The power prediction system forms the fitting value of the second future power generation data set in the ultra-short-term prediction period according to the fitting values ​​of the 16 corresponding second future power generation data and the second moment data corresponding to the fitting value of the second future power generation data.

[0185] For the ultra-short-term forecast period, the above R H By replacing it with NWP_DQ, the fitting value of the second future power generation data in the short-term prediction period (also recorded as f1) can be obtained. The power prediction system forms the fitting value of the second future power generation data set in the short-term prediction period based on the fitting value of the second future power generation data and the second moment data corresponding to the fitting value of the second future power generation data.

[0186] When the power prediction system determines that the second prediction data table and the third prediction data table do not satisfy the first preset correction condition, the power prediction system does not execute step S143 and step S144.

[0187] In step S144, the power prediction system updates the second future power generation data set according to the fitting value of the second future power generation data set and the second future power generation data set.

[0188] According to an example embodiment, the power prediction system may update the second future power generation data set based on the fitting value of the second future power generation data set, the regression method to correct the first-order fitting coefficient, and the second future power generation data set.

[0189] For example, the power prediction system can compare the second future power generation data in step S142 with the fitting value of the second future power generation data × the regression method corrected first fitting coefficient. The regression method corrected first fitting coefficient can be 0.8. The power prediction system takes the maximum value of the two as the updated second future power generation data (the updated second future power generation data is also recorded as UST_F2), thereby updating the second future power generation data set.

[0190] Alternatively, see Figure 6 , step S140 may further include step S145.

[0191] In step S145, when the second prediction data table and the third prediction data table meet the second preset correction condition, the power prediction system processes the second prediction data table and the third prediction data table according to the eighth preset processing rule to update the second future power generation data set.

[0192] The second preset correction condition is a condition for performing sunny day correction and sunrise and sunset correction on the second future power generation data set. The second preset correction condition may include a preset all-sunny day correction condition and a preset super-sunny day correction condition.

[0193] For example, the preset all-sunny day correction condition may be: the power prediction system determines that sumnwp_now≥sumnwp×all-sunny day allocation ratio of the regression method (the all-sunny day allocation ratio of the regression method may be 1.2) and max(UST_F2 in step S142 or UST_F2 in step S144)<installed capacity×0.75. Then the power prediction system takes the updated second future power generation data as UST_F2=(UST_F2 in step S142 or UST_F2 in step S144)×1.2.

[0194] The preset super-sunny correction condition may be: the power prediction system determines that sumnwp_now≥sumnwp×regression method super-sunny allocation ratio (regression method super-sunny allocation ratio may be 1.6) and max(UST_F2 in step S142 or UST_F2 in step S144)<installed capacity×0.75. Then the power prediction system takes the updated second future power generation data as UST_F2=(UST_F2 in step S142 or UST_F2 in step S144)×1.3.

[0195] When the power prediction system determines that the second prediction data table and the third prediction data table meet the second preset correction condition, the power prediction system performs data processing on the second prediction data table and the third prediction data table according to the eighth preset processing rule.

[0196] The eighth preset processing rule is a preset rule for performing data processing on the second prediction data table and the third prediction data table.

[0197] For example, the eighth preset processing rule may be:

[0198] 1. Calculate the average value of the actual power generation data in the second prediction data table according to the data at the second moment, and obtain the average value of the actual power generation data at each second moment (denoted as rpmean4, the number of rpmean4 should be NN).

[0199] 2. Add 18 zeros before rpmean4 and 5 zeros after it to form 96 actual power generation data points for one day (recorded as rpmean5). The 96 actual power generation data points all correspond to the time data one by one.

[0200] For the ultra-short-term prediction period, the power prediction system takes out the actual power generation data corresponding to the second moment data from rpmean5 and obtains 16 corresponding second future power generation data; among them, if a moment is before 8:30 am or after 4:00 pm, then the updated second future power generation data at that moment is the value corresponding to that moment in rpmean5 × 0.5 + the value corresponding to that moment in UST_F2 × 0.5 (where 0.5 is a coefficient, which can be adjusted according to the natural environment of the distributed photovoltaic power station); and the future predicted irradiance data (R H ) is 0, the updated second future power generation data is also 0.

[0201] For the short-term prediction time period, the power prediction system takes out the actual power generation data corresponding to the second moment data from rpmean5, and obtains 96×7 corresponding second future power generation data; among them, if a certain moment is before 8:30 in the morning or after 4:00 in the afternoon, then the updated second future power generation data at that moment is the value corresponding to that moment in rpmean5 × 0.5 + the value corresponding to that moment in UST_F2 × 0.5 (where 0.5 is a coefficient, which can be adjusted according to the natural environment of the distributed photovoltaic power station); and when the future predicted irradiance data (NWP_DQ) is 0, the updated second future power generation data is also 0.

[0202] The second future generated power data updated in step S145 is also recorded as UST_F2.

[0203] When the power prediction system determines that the second prediction data table and the third prediction data table do not satisfy the second preset correction condition, step S145 is not executed.

[0204] Alternatively, see Figure 7 , step S150 may further include step S151.

[0205] In step S145, the power prediction system determines a comprehensive future power generation data set according to the first future power generation data set and the corresponding coefficient, the second future power generation data set and the corresponding coefficient.

[0206] For example, the power forecasting system can calculate the comprehensive future power generation data according to the following formula:

[0207] UST=0.5×UST_F1+0.5×UST_F2;

[0208] Among them, UST is the comprehensive future power generation data; UST_F2 can be UST_F2 in step S142, UST_F2 in step S144 or UST_F2 in step S145.

[0209] The power prediction system can form a comprehensive future power generation data set based on the comprehensive future power generation data and the second moment data corresponding to the comprehensive future power generation data.

[0210] According to another aspect of the present application, the present application provides a power prediction system 200 for a distributed photovoltaic power station. The power prediction system 200 is used to execute the power prediction method 1000 for a distributed photovoltaic power station as described above. Figure 8 , the power prediction system 200 includes a processing unit 210 .

[0211] The processing unit 210 determines a historical data set of the distributed photovoltaic power station in the first preset time period based on the start time, and the historical data set includes historical predicted irradiance data and actual power generation data determined according to the first preset time period, and first moment data corresponding to the historical predicted irradiance data and the actual power generation data.

[0212] The processing unit 210 determines a future irradiance data set of the distributed photovoltaic power station in a second preset time period based on the start time, and the future irradiance data set includes future predicted irradiance data determined according to the second preset time period and second moment data corresponding to the future predicted irradiance data.

[0213] The processing unit 210 determines a first future power generation data set for a second preset time period based on a first preset algorithm according to the historical data set and the future irradiance data set. The first future power generation data set includes first future power generation data corresponding to each second preset time period and second moment data corresponding to the first future power generation data.

[0214] The processing unit 210 determines a second future power generation data set for a second preset time period based on a second preset algorithm according to the historical data set and the future irradiance data set. The second future power generation data set includes second future power generation data corresponding to each second preset time period and second moment data corresponding to the second future power generation data.

[0215] The processing unit 210 determines a comprehensive future power generation data set of the distributed photovoltaic power station based on the first future power generation data set and the second future power generation data set. The comprehensive future power generation data set includes comprehensive predicted power generation data corresponding to each second preset time period and second moment data corresponding to the comprehensive predicted power generation data.

[0216] According to an example embodiment, the start time, the first preset time period, the historical data set, the first preset time period, the historical predicted irradiance data, the actual power generation data, the first moment data, the second preset time period, the future irradiance data, the second preset time period, the future predicted irradiance data, the first preset algorithm, the first future power generation data set, the second preset algorithm, the second future power generation data set and the comprehensive future power generation data set have been described in the power prediction method 1000 above and will not be repeated here.

[0217] This application can predict the power generation of distributed photovoltaic power stations through historical predicted irradiance data, actual power generation data and future irradiance data sets. This application can predict the power generation of distributed photovoltaic power stations through two algorithms, which can improve the accuracy of prediction, reduce the uncertainty of photovoltaic power generation, and improve the ability of the power grid to accept random fluctuating power sources.

[0218] According to another aspect of the present application, the present application further provides a non-volatile computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the power prediction method of the distributed photovoltaic power station as described above can be implemented.

[0219] According to another aspect of the present application, the present application also provides an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the one or more processors are enabled to implement the power prediction method of the distributed photovoltaic power station as described above.

[0220] According to another aspect of the present application, the present application also provides a computer program product, including: a computer program stored on a computer-readable storage medium; the computer program includes program instructions, and when the program instructions are executed by the computer, the computer executes the power prediction method of the distributed photovoltaic power station as described above.

[0221] Finally, it should be noted that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Although the present application is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions of the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A power prediction method for a distributed photovoltaic power station, characterized in that: The power prediction method comprises: Based on the start time, determine a historical data set of the distributed photovoltaic power station in a first preset time period, the historical data set including historical predicted irradiance data, actual power generation data determined according to the first preset time period, and first moment data corresponding to the historical predicted irradiance data and the actual power generation data; Based on the start time, determining a future irradiance data set of the distributed photovoltaic power station in a second preset time period, the future irradiance data set comprising future predicted irradiance data determined according to the second preset time period, and second time point data corresponding to the future predicted irradiance data; Determine, based on the historical data set and the future irradiance data set, a first future power generation data set for the second preset time period based on a first preset algorithm, wherein the first future power generation data set includes first future power generation data corresponding to each of the second preset time periods, and the second moment data corresponding to the first future power generation data; Determine, based on the historical data set and the future irradiance data set, a second future power generation data set for the second preset time period based on a second preset algorithm, wherein the second future power generation data set includes second future power generation data corresponding to each second preset time period, and the second moment data corresponding to the second future power generation data; Based on the first future power generation data set and the second future power generation data set, a comprehensive future power generation data set of the distributed photovoltaic power station is determined, and the comprehensive future power generation data set includes the comprehensive future predicted power generation data corresponding to each of the second preset time periods, and the second moment data corresponding to the comprehensive future predicted power generation data.

2. The power prediction method according to claim 1, characterized in that: The step of determining the historical data set of the distributed photovoltaic power station in the first preset time period based on the start time includes: Based on the start time, determining an initial historical data set of the distributed photovoltaic power station in the first preset time period; The initial historical data set is processed according to a first preset processing rule to obtain the historical data set.

3. The power prediction method according to claim 1, characterized in that: The determining, based on the historical data set and the future irradiance data set and based on a first preset algorithm, a first future power generation data set for the second preset time period includes: Determine a first prediction data table according to the historical data set; According to a second preset processing rule, data processing is performed on the first prediction data table to obtain a first component of the first future power generation data set; According to a third preset processing rule, data processing is performed on the historical data set and the first prediction data table to obtain a second component of the first future power generation data set; The first future generated power data set is determined according to the first component and the second component.

4. The power prediction method according to claim 3, characterized in that: After determining the first future power generation data set according to the first component and the second component, the power prediction method further includes: When the historical data set and the first prediction data table satisfy a first preset extreme weather condition, data processing is performed on the historical data set and the first prediction data table according to a fourth preset processing rule to update the first future power generation data set.

5. The power prediction method according to claim 3, characterized in that: After determining the first future power generation data set according to the first component and the second component, the power prediction method further includes: When the historical data set and the first prediction data table satisfy a second preset extreme weather condition, data processing is performed on the historical data set and the first prediction data table according to a fifth preset processing rule to update the first future power generation data set.

6. The power prediction method according to claim 1, characterized in that: The determining, based on the historical data set and the future irradiance data set and based on a second preset algorithm, a second future power generation data set for the second preset time period includes: Determine a second prediction data table and a third prediction data table according to the historical data set; According to a sixth preset processing rule, data processing is performed on the second prediction data table and the third prediction data table to obtain the second future generated power data set.

7. The power prediction method according to claim 6, characterized in that: After performing data processing on the second prediction data table and the third prediction data table according to a sixth preset processing rule to obtain the second future generated power data set, the power control method further includes: When the second prediction data table and the third prediction data table meet the first preset correction condition, data processing is performed on the second prediction data table and the third prediction data table according to a seventh preset processing rule to obtain a fitting value of the second future generated power data set; The second future power generation data set is updated according to the fitting value of the second future power generation data set and the second future power generation data set.

8. The power prediction method according to claim 7, characterized in that: After updating the second future power generation data set according to the fitted value of the second future power generation data set and the second future power generation data set, the power prediction method further includes: When the second prediction data table and the third prediction data table satisfy the second preset correction condition, data processing is performed on the second prediction data table and the third prediction data table according to an eighth preset processing rule to update the second future power generation data set.

9. The power prediction method according to claim 1, characterized in that: Determining the comprehensive future power generation data set of the distributed photovoltaic power station according to the first future power generation data set and the second future power generation data set includes: The comprehensive future power generation data set is determined according to the first future power generation data set and the corresponding coefficient, the second future power generation data set and the corresponding coefficient.

10. A power prediction system for a distributed photovoltaic power station, characterized in that: The power prediction system is used to execute the power prediction method of the distributed photovoltaic power station according to any one of claims 1 to 9, and the power prediction system includes: A processing unit, based on the start time, determines a historical data set of the distributed photovoltaic power station in a first preset time period, wherein the historical data set includes historical predicted irradiance data, actual power generation data, and first moment data corresponding to the historical predicted irradiance data and the actual power generation data determined according to the first preset time period; The processing unit determines, based on the start time, a future irradiance data set of the distributed photovoltaic power station in a second preset time period, wherein the future irradiance data set includes future predicted irradiance data determined according to the second preset time period, and second time point data corresponding to the future predicted irradiance data; The processing unit determines, based on the first preset algorithm and according to the historical data set and the future irradiance data set, a first future power generation data set for the second preset time period, wherein the first future power generation data set includes first future power generation data corresponding to each of the second preset time periods, and the second moment data corresponding to the first future power generation data; The processing unit determines, based on the second preset algorithm and according to the historical data set and the future irradiance data set, a second future power generation data set for the second preset time period, wherein the second future power generation data set includes second future power generation data corresponding to each second preset time period and the second moment data corresponding to the second future power generation data; The processing unit determines a comprehensive future power generation data set of the distributed photovoltaic power station based on the first future power generation data set and the second future power generation data set, and the comprehensive future power generation data set includes the comprehensive predicted power generation data corresponding to each of the second preset time periods, and the second moment data corresponding to the comprehensive predicted power generation data.

Images