Distributed photovoltaic ultra-short-term power prediction method and device based on remote sensing information

Abstract

The invention provides a distributed photovoltaic ultra-short-term power prediction method and device based on remote sensing information, and the method comprises the steps: obtaining a distributed photovoltaic ultra-short-term power prediction model through a cloud condition power change relational expression and a solar resource utilization rate relational expression according to the remote sensing information of a target satellite, a to-be-predicted time period and basic information; generating distributed photovoltaic power variation and distributed photovoltaic theoretical power of the to-be-predicted time period; through an ultra-short-term power prediction model, according to target meteorological monitoring data, target satellite remote sensing information, distributed photovoltaic power variation and distributed photovoltaic theoretical power of a to-be-predicted time period, distributed photovoltaic predicted power of the to-be-predicted time period is generated, and the ultra-short-term power prediction model is obtained by training based on a dynamic recurrent neural network. Satellite remote sensing information and real-time meteorological monitoring data of an automatic meteorological station are combined, ultra-short-term power prediction of a distributed photovoltaic power generation cluster is achieved, the utilization efficiency of green energy is improved to the maximum extent, and the safety of a power grid is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a distributed photovoltaic ultra-short-term power prediction method and device based on remote sensing information. Background Art

[0002] Photovoltaic power prediction is a key technology in solar photovoltaic power generation systems. It is mainly used to predict the power output of photovoltaic power stations in the future. Accurate photovoltaic power prediction can ensure the stability of power supply and alleviate the problem of imbalance between power supply and demand. In related technologies, the ultra-short-term power prediction of photovoltaic power stations is based on the telemetry data of automatic weather stations, and the optimal statistical method is used for modeling to predict the power generation power in the next 0-4 hours. This method has achieved the prediction of ultra-short-term photovoltaic power station power to a certain extent. However, for regional distributed photovoltaic power generation systems with wide spatial coverage, low single-unit capacity, flexible and diverse configuration forms, and significant differences in power generation characteristics, as well as distributed photovoltaic power generation clusters composed of several distributed photovoltaic power generation systems in a certain area, the accuracy of photovoltaic power prediction results of conventional photovoltaic power generation systems is poor, which cannot guarantee the safety of the power grid, and the utilization efficiency of photovoltaic green energy is low. Summary of the invention

[0003] One object of the present invention is to provide a distributed photovoltaic ultra-short-term power prediction method based on remote sensing information, combining satellite remote sensing information with real-time meteorological monitoring data from automatic weather stations, to achieve ultra-short-term power prediction of a large number of distributed photovoltaic power generation clusters within a large spatial range, based on solving the power generation and grid-connected scheduling problems of photovoltaic and other new energy sources, maximizing the utilization efficiency of green energy, ensuring the safety of the power grid, and providing certain technical support for the safe and stable operation of the distribution network under small load and large power supply scenarios. Another object of the present invention is to provide a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information. Another object of the present invention is to provide a computer-readable medium. Another object of the present invention is to provide a computer device.

[0004] In order to achieve the above objectives, the present invention discloses a distributed photovoltaic ultra-short-term power prediction method based on remote sensing information, comprising:

[0005] Obtain basic information of distributed photovoltaic power generation systems within a specified time period, as well as target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation systems are located;

[0006] The distributed photovoltaic power change is generated based on the pre-constructed cloud condition power change relationship and the target satellite remote sensing information;

[0007] Through the pre-built solar energy resource utilization relationship, according to the forecast period and basic information, the distributed photovoltaic theoretical power of the forecast period is generated;

[0008] Through the pre-constructed ultra-short-term power prediction model, the distributed photovoltaic predicted power for the period to be predicted is generated according to the target meteorological monitoring data, target satellite remote sensing information, distributed photovoltaic power change and distributed photovoltaic theoretical power for the period to be predicted. The ultra-short-term power prediction model is trained based on a dynamic recursive neural network.

[0009] Preferably, the method further comprises:

[0010] Obtain the historical active power and operating status data of the distributed photovoltaic power generation system, the historical satellite remote sensing information and historical meteorological monitoring data in the area where the distributed photovoltaic power generation system is located;

[0011] Generate distributed photovoltaic historical power time series data according to the preset time resolution, based on basic information, operating status data and historical active power;

[0012] According to the distributed photovoltaic historical power time series data, the historical satellite remote sensing information and historical meteorological monitoring data are synchronized to generate historical satellite remote sensing time series data and historical meteorological monitoring time series data;

[0013] Based on basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data, cloud condition power change relationship and solar energy resource utilization relationship are constructed.

[0014] Preferably, based on basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data, cloud condition power change relationship and solar energy resource utilization relationship are constructed, including:

[0015] Extract effective data from distributed photovoltaic historical power time series data based on basic information to determine distributed photovoltaic historical effective power data;

[0016] According to the target period corresponding to the historical effective power data of distributed photovoltaics, the historical satellite remote sensing data and the historical meteorological monitoring data for the same period are extracted from the historical satellite remote sensing time series data and the historical meteorological monitoring time series data;

[0017] Based on the historical satellite remote sensing data and the historical effective power data of distributed photovoltaics, the cloud condition power change relationship is constructed;

[0018] Based on the historical satellite remote sensing data, historical meteorological monitoring data and historical effective power data of distributed photovoltaics for the same period, a solar energy resource utilization rate relationship is constructed.

[0019] Preferably, the satellite remote sensing data for the same period in history include cloud cover for the same period in history;

[0020] Based on the historical satellite remote sensing data and distributed photovoltaic historical effective power data of the same period, the cloud power change relationship is constructed, including:

[0021] Determine the non-zero cloud cover period based on the historical cloud cover for the same period;

[0022] Filter out the historical cloud cover corresponding to the non-zero cloud cover period from the historical cloud cover of the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the non-zero cloud cover period from the distributed photovoltaic historical effective power data;

[0023] According to the preset cloud amount level mapping relationship, the historical cloud amounts corresponding to the non-zero cloud amount period are classified into levels, and the cloud amount level corresponding to each non-zero cloud amount period is generated;

[0024] According to the cloud level corresponding to each non-zero cloud period and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud period, a cloud condition power linear regression model is constructed;

[0025] The least square method is used to solve the parameters of the cloud power linear regression model and generate the cloud power variation relationship.

[0026] Preferably, the satellite remote sensing data for the same period in history include cloud cover for the same period in history, and the meteorological monitoring data for the same period in history include total solar radiation data for the same period in history;

[0027] Based on the historical satellite remote sensing data, historical meteorological monitoring data and distributed photovoltaic historical effective power data, a solar energy resource utilization rate relationship is constructed, including:

[0028] Determine the zero cloud period based on the historical cloud cover during the same period;

[0029] Filter out the total solar radiation data corresponding to the zero cloud period from the historical total solar radiation data for the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the zero cloud period from the distributed photovoltaic historical effective power data;

[0030] According to the total solar radiation data corresponding to the zero cloud period and the historical distributed photovoltaic effective power data corresponding to the zero cloud period, a linear regression model of solar energy resource utilization is constructed;

[0031] The least square method is used to solve the parameters of the solar energy resource utilization linear regression model and generate a solar energy resource utilization relationship.

[0032] Preferably, the basic information includes the location latitude of the distributed photovoltaic power generation system;

[0033] Through the pre-built solar energy resource utilization relationship, according to the forecast period and basic information, the distributed photovoltaic theoretical power of the forecast period is generated, including:

[0034] Generate the solar declination, hour angle and the difference in distance between the sun and the earth according to the period to be predicted;

[0035] Generate the cosine value of the solar zenith angle according to the latitude of the distributed photovoltaic power generation system, the solar declination and the hour angle;

[0036] The solar shortwave radiation at the upper boundary of the atmosphere is generated based on the pre-acquired solar constant, the cosine value of the solar zenith angle, the latitude of the distributed photovoltaic power generation system, the solar declination, the hour angle and the difference in the distance between the sun and the earth;

[0037] Through the solar energy resource utilization rate relationship, the distributed photovoltaic theoretical power for the predicted period is generated according to the solar shortwave radiation at the upper boundary of the atmosphere.

[0038] The present invention also discloses a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information, comprising:

[0039] A prediction data acquisition unit, used to acquire basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located;

[0040] A power variation generating unit is used to generate a distributed photovoltaic power variation according to the target satellite remote sensing information through a pre-built cloud condition power variation relationship;

[0041] The theoretical power generation unit is used to generate the distributed photovoltaic theoretical power of the time period to be predicted according to the time period to be predicted and the basic information through the pre-built solar energy resource utilization rate relationship;

[0042] The ultra-short-term power prediction unit is used to generate the distributed photovoltaic predicted power for the time period to be predicted based on the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted through a pre-built ultra-short-term power prediction model. The ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network.

[0043] The present invention also discloses a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0044] The present invention also discloses a computer device, including a memory and a processor, wherein the memory is used to store information including program instructions, the processor is used to control the execution of program instructions, and the processor implements the above method when executing the program.

[0045] The present invention also discloses a computer program product, including a computer program / instruction, and the method described above is implemented when the computer program / instruction is executed by a processor.

[0046] The present invention obtains basic information of a distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in an area where the distributed photovoltaic power generation system is located; generates a distributed photovoltaic power change according to target satellite remote sensing information through a pre-constructed cloud condition power change relationship; generates a distributed photovoltaic theoretical power for a time period to be predicted according to a time period to be predicted and basic information through a pre-constructed solar energy resource utilization relationship; generates a distributed photovoltaic predicted power for a time period to be predicted according to target meteorological monitoring data, target satellite remote sensing information, distributed photovoltaic power change and distributed photovoltaic theoretical power for a time period to be predicted through a pre-constructed ultra-short-term power prediction model, wherein the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network, and combines satellite remote sensing information with real-time meteorological monitoring data of an automatic weather station to realize ultra-short-term power prediction of a large number of distributed photovoltaic power generation clusters within a large spatial range, based on solving the problem of power generation and grid-connected scheduling of new energy such as photovoltaics, maximizing the utilization efficiency of green energy, ensuring the safety of the power grid, and providing certain technical support for the safe and stable operation of the distribution network under small load and large power supply scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A flow chart of a distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention;

[0049] Figure 2 A flowchart of another distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of the structure of a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information provided by an embodiment of the present invention;

[0051] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] In order to facilitate the understanding of the technical solution provided by this application, the relevant contents of the technical solution of this application are first explained below. The ultra-short-term photovoltaic power station power prediction system can effectively solve the negative impact of intermittent and volatile power generation of new energy caused by significant changes in the sky cloud system during the period of concentrated grid-connected power generation of regional distributed photovoltaic clusters, and correct the power generation plan of the power grid dispatching department in real time to ensure the safety of the power grid and maximize the utilization efficiency of photovoltaic green energy. For regional distributed photovoltaic power generation systems with wide spatial coverage, low single-unit capacity, flexible and diverse configuration forms, and significant differences in power generation characteristics, as well as distributed photovoltaic power generation clusters composed of several distributed photovoltaic power generation systems in a certain area, their ultra-short-term power prediction needs to introduce meteorological data that matches the coverage range and is stable and reliable. Satellite cloud images can capture a wide range of water vapor, visible light and cloud cover data, monitor sensitive factors that affect solar radiation elements in real time (such as cloud cover, cloud type, cloud thickness, cloud top height, etc.), and track the movement path and trend of weather systems in a timely manner (such as cloud track wind technology). It is a key information that can effectively improve the power prediction of ultra-short-term photovoltaic power stations. In summary, the ultra-short-term power prediction based on satellite remote sensing and meteorological monitoring data can improve the traditional power prediction algorithm framework and use effective data to improve the accuracy of regional distributed photovoltaic power prediction. The present invention realizes regional distributed photovoltaic ultra-short-term power prediction based on the historical operation information of regional distributed photovoltaic power generation systems and historical satellite remote sensing data and meteorological monitoring data of the same period, providing a reference for operation and management for power grid dispatching and control agencies or photovoltaic power generation enterprises.

[0054] The following takes a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information as an example of an execution subject to illustrate the implementation process of the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention. It can be understood that the execution subject of the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention includes but is not limited to a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information.

[0055] Figure 1 A flowchart of a distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method includes:

[0056] Step 101: Obtain basic information of a distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in an area where the distributed photovoltaic power generation system is located.

[0057] Step 102: Generate a distributed photovoltaic power change amount according to the target satellite remote sensing information through the pre-constructed cloud condition power change relationship.

[0058] In the embodiment of the present invention, a cloud condition power variation relationship is constructed based on basic information, distributed photovoltaic historical power time series data and historical satellite remote sensing time series data.

[0059] Step 103: Generate the distributed photovoltaic theoretical power for the time period to be predicted based on the time period to be predicted and basic information through the pre-constructed solar energy resource utilization relationship.

[0060] In the embodiment of the present invention, a solar energy resource utilization rate relational expression is constructed based on basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data.

[0061] Step 104: Generate the distributed photovoltaic predicted power for the time period to be predicted by using the pre-built ultra-short-term power prediction model according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted.

[0062] In the embodiment of the present invention, the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network.

[0063] In the technical solution provided by the embodiment of the present invention, basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located are obtained; a distributed photovoltaic power change amount is generated according to the target satellite remote sensing information through a pre-constructed cloud condition power change relationship; a distributed photovoltaic theoretical power of the time period to be predicted is generated according to the time period to be predicted and the basic information through a pre-constructed solar energy resource utilization relationship; a distributed photovoltaic predicted power of the time period to be predicted is generated according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change amount and the distributed photovoltaic theoretical power of the time period to be predicted through a pre-constructed ultra-short-term power prediction model, and the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network, and combines satellite remote sensing information with real-time meteorological monitoring data of an automatic weather station to realize ultra-short-term power prediction of a large number of distributed photovoltaic power generation clusters within a large spatial range, based on solving the problem of power generation and grid-connected scheduling of new energy such as photovoltaics, maximizing the utilization efficiency of green energy, ensuring the safety of the power grid, and providing certain technical support for the safe and stable operation of the distribution network under small load and large power supply scenarios.

[0064] Figure 2 A flowchart of another distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by an embodiment of the present invention is as follows: Figure 2 As shown, the method includes:

[0065] Step 201: Obtain historical active power and operating status data of a distributed photovoltaic power generation system, historical satellite remote sensing information and historical meteorological monitoring data in the area where the distributed photovoltaic power generation system is located.

[0066] In the embodiment of the present invention, each step is performed by a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information.

[0067] In an embodiment of the present invention, basic information, historical active power, and operating status data of distributed photovoltaic power generation systems in the intended calculation area in a preset historical time period are obtained. Among them, the basic information includes but is not limited to the rated installed capacity, access voltage level, access point, grid connection time, latitude of the distributed photovoltaic power generation system location, and longitude of the distributed photovoltaic power generation system location; the operating status data includes but is not limited to normal or faulty; historical satellite remote sensing information includes but is not limited to gridded cloud data; historical meteorological monitoring data includes monitoring data of multiple meteorological elements, and meteorological elements include but are not limited to temperature, total solar radiation, wind speed, and relative humidity, and historical meteorological monitoring data includes temperature monitoring data, total solar radiation monitoring data, wind speed monitoring data, and relative humidity monitoring data.

[0068] It is worth noting that the historical time period can be set according to actual needs, and the embodiment of the present invention does not limit this.

[0069] Step 202: Generate distributed photovoltaic historical power time series data according to the preset time resolution, basic information, operating status data and historical active power.

[0070] In the embodiment of the present invention, based on the operating status data, historical active power with normal operating status data is screened out from historical active power, thereby improving the accuracy of subsequent training models.

[0071] Specifically, according to the time resolution, the distributed photovoltaic historical power time series data is generated based on the basic information and operation status data of each photovoltaic power generation system as normal historical active power, wherein the basic information is used as the attribute information of each photovoltaic power generation system.

[0072] It is worth noting that the time resolution can be set according to actual needs, and the embodiment of the present invention does not limit this. As an optional solution, the time resolution is 15 minutes.

[0073] Step 203: synchronize the historical satellite remote sensing information and the historical meteorological monitoring data according to the distributed photovoltaic historical power time series data, and generate the historical satellite remote sensing time series data and the historical meteorological monitoring time series data.

[0074] Specifically, the stationary historical satellite remote sensing information is selected and clock-synchronized with the distributed photovoltaic historical power time series data. The gridded cloud data provided by the satellite is organized into historical satellite remote sensing time series data. The historical meteorological monitoring data is synchronized with the distributed photovoltaic historical power time series data to generate historical meteorological monitoring time series data.

[0075] In the embodiment of the present invention, the time resolutions of the distributed photovoltaic historical power time series data, the historical satellite remote sensing time series data and the historical meteorological monitoring time series data are the same.

[0076] Step 204: extract effective data from the distributed photovoltaic historical power time series data according to the basic information to determine the distributed photovoltaic historical effective power data.

[0077] In an embodiment of the present invention, the basic information includes the rated installed capacity, which specifically refers to the maximum output power of a photovoltaic module or system under standard test conditions; the distributed photovoltaic historical power time series data includes multiple historical time periods, the historical time point corresponding to each historical time period, and the historical power corresponding to the historical time point.

[0078] Specifically, the historical power corresponding to each historical time point is compared with the rated installed capacity to determine whether the historical power is greater than the preset abnormal power threshold. If so, it indicates that the adjacent historical time period corresponding to the historical time point is a period of abnormal change in distributed photovoltaics, and the data in this period is valid data. Based on the time range of the distributed photovoltaic abnormal change event, the distributed photovoltaic historical power time series data corresponding to the adjacent historical time period is extracted, and the extracted data is determined as the distributed photovoltaic historical effective power data; if not, it indicates that the adjacent historical time period corresponding to the historical time point is a period of normal change in distributed photovoltaics, and no abnormal impact has occurred.

[0079] It is worth noting that the abnormal power threshold can be set according to actual needs, and this is not limited in the embodiment of the present invention. As an optional solution, the abnormal power threshold is 1.3 times the rated installed capacity.

[0080] Step 205: according to the target period corresponding to the historical effective power data of distributed photovoltaics, extract the historical satellite remote sensing data and the historical meteorological monitoring data for the same period from the historical satellite remote sensing time series data and the historical meteorological monitoring time series data.

[0081] In the embodiment of the present invention, according to the target period corresponding to the historical effective power data of distributed photovoltaic, the historical satellite remote sensing data and the historical meteorological monitoring data of the corresponding period are extracted. The historical satellite remote sensing data of the corresponding period include the historical cloud cover, that is, the gridded cloud cover data corresponding to the target period; the historical meteorological monitoring data of the corresponding period include the temperature monitoring data, the total solar radiation monitoring data, the wind speed monitoring data and the relative humidity monitoring data corresponding to the target period.

[0082] Step 206: construct a cloud power change relationship based on the historical satellite remote sensing data and the distributed photovoltaic historical effective power data for the same period.

[0083] In the embodiment of the present invention, the satellite remote sensing data for the same period in history includes the cloud cover for the same period in history, that is, the gridded cloud cover data corresponding to the target time period.

[0084] In the embodiment of the present invention, step 206 specifically includes:

[0085] Step 2061: Determine a non-zero cloud cover period based on the historical cloud cover for the same period.

[0086] In the embodiment of the present invention, the historical cloud cover for the same period includes gridded cloud cover data corresponding to the target time period. According to the cloud cover data, the time period in which the cloud cover is not zero is screened out and determined as the non-zero cloud cover time period.

[0087] Step 2062: Filter out the historical cloud cover corresponding to the non-zero cloud cover period from the historical cloud cover of the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the non-zero cloud cover period from the distributed photovoltaic historical effective power data.

[0088] Specifically, according to the non-zero cloud cover period, the corresponding historical cloud cover for the same period is screened out; according to the non-zero cloud cover period, the corresponding distributed photovoltaic historical effective power data is determined.

[0089] Step 2063: According to the preset cloud cover level mapping relationship, the historical cloud cover corresponding to the non-zero cloud cover period is classified into levels, and the cloud cover level corresponding to each non-zero cloud cover period is generated.

[0090] In the embodiment of the present invention, the cloud level mapping relationship includes cloud data and corresponding cloud levels, and the cloud levels include levels 1 to 10. It is worth noting that the cloud level mapping relationship can be set according to actual needs, and the embodiment of the present invention does not limit this.

[0091] Specifically, the cloud amount data of the historical cloud amount corresponding to each non-zero cloud amount period is matched with the cloud amount data in the cloud amount level mapping relationship to determine the cloud amount level corresponding to each non-zero cloud amount period.

[0092] Step 2064: construct a cloud condition power linear regression model based on the cloud level corresponding to each non-zero cloud period and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud period.

[0093] In an embodiment of the present invention, a linear regression model is used to calculate the cloud level corresponding to each non-zero cloud period and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud period, and the correlation between different levels of cloud cover and the changes in active power of the distributed photovoltaic system (or cluster) during the same period is statistically analyzed to construct a cloud condition power linear regression model.

[0094] Step 2065: Use the least squares method to solve the parameters of the cloud condition power linear regression model and generate a cloud condition power change relationship.

[0095] Specifically, the linear parameters of the cloud condition power linear regression model are solved by the least squares method, and the cloud condition power change relationship is constructed based on the linear parameters. The cloud condition power change relationship is used to describe the correlation between different levels of cloud cover and the active power change of the distributed photovoltaic system (or cluster) in the same period.

[0096] Step 207: construct a solar energy resource utilization rate relationship based on historical satellite remote sensing data, historical meteorological monitoring data and distributed photovoltaic historical effective power data for the same period.

[0097] In the embodiment of the present invention, the satellite remote sensing data for the same period in history include cloud cover for the same period in history, and the meteorological monitoring data for the same period in history include total solar radiation data for the same period in history.

[0098] In the embodiment of the present invention, step 207 specifically includes:

[0099] Step 2071: Determine the zero cloud period based on the historical cloud cover during the same period.

[0100] In the embodiment of the present invention, the historical cloud cover for the same period includes gridded cloud cover data corresponding to the target time period. According to the cloud cover data, the time period with a cloud cover of 0 is screened out and determined as the zero cloud cover period.

[0101] Step 2072: Filter out the total solar radiation data corresponding to the zero cloud period from the historical total solar radiation data for the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the zero cloud period from the distributed photovoltaic historical effective power data.

[0102] Specifically, according to the zero cloud period, the corresponding total solar radiation data is screened out; according to the zero cloud period, the corresponding distributed photovoltaic historical effective power data is determined.

[0103] Step 2073: construct a solar energy resource utilization linear regression model based on the total solar radiation data corresponding to the zero cloud period and the distributed photovoltaic historical effective power data corresponding to the zero cloud period.

[0104] In an embodiment of the present invention, a linear regression model is used to calculate the total solar radiation data corresponding to each zero cloud period and the distributed photovoltaic historical effective power data corresponding to the zero cloud period, and the correlation between the total solar radiation data of the automatic weather station and the actual value of the active power in the same period is statistically analyzed to construct a linear regression model for solar energy resource utilization.

[0105] Step 2074: Solve the parameters of the solar energy resource utilization linear regression model by the least square method to generate a solar energy resource utilization relationship.

[0106] Specifically, the linear parameters of the linear regression model of solar energy resource utilization rate are solved by the least square method, and the solar energy resource utilization rate relationship is constructed according to the linear parameters. The solar energy resource utilization rate relationship is used to describe the correlation between the total solar radiation data and the actual value of active power in the same period.

[0107] Step 208: construct an ultra-short-term power prediction model.

[0108] In the embodiment of the present invention, the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network. As an optional solution, the dynamic recursive neural network is an Elman neural network.

[0109] In the embodiment of the present invention, the daily variation curve of the solar shortwave radiation irradiance at the upper boundary of the atmosphere in the northern hemisphere is calculated according to the daily variation law of the solar altitude angle at the top of the atmosphere. The calculation method is as follows: let the solar shortwave radiation received per unit area at any position at the upper boundary of the atmosphere in the northern hemisphere be S (ignoring the influence of the distance from the sun to the upper boundary of the atmosphere caused by the change in latitude), through S=S0×sin h, sin h=sinαsinθ+sinαcosθcost, we get: S=S0sinα(sinθ+cosθcost). Wherein, S is the solar shortwave radiation; S0 is the solar constant, and S0 is taken as 1368W / m 2 ; h is the solar altitude angle; θ is the solar declination; t is the hour angle; α is the latitude of the distributed photovoltaic power generation system.

[0110] If we consider the difference in the distance between the sun and the earth on different dates:

[0111]

[0112] θ0=(2πN-1) / 365

[0113] Among them, θ0 is the latitude of the sun's direct point, N is the day of the year, is the difference in distance between the Earth and the Sun.

[0114] Then the solar radiation per unit area reaching any position at the upper boundary of the northern hemisphere atmosphere can be written as:

[0115]

[0116] Where S is the solar shortwave radiation; S0 is the solar constant, S0 is 1368W / m 2 ; θ is the solar declination; t is the hour angle; α is the latitude of the distributed photovoltaic power generation system; is the average distance between the sun and the earth; d is the distance between the sun and the earth; cosθ sun is the cosine of the solar zenith angle.

[0117] Specifically, the solar shortwave radiation corresponding to each target period is calculated by the above formula; the solar shortwave radiation corresponding to each target period is calculated by the solar energy resource utilization rate relationship to generate the distributed photovoltaic theoretical power corresponding to each target period.

[0118] In the embodiment of the present invention, step 208 is specifically as follows: according to the distributed photovoltaic theoretical power corresponding to each target time period, the distributed photovoltaic historical effective power data, the historical satellite remote sensing data and the historical meteorological monitoring data of the same period, the Elman neural network is iteratively trained to construct an ultra-short-term power prediction model.

[0119] In the embodiment of the present invention, the delay and storage characteristics of the Elman neural network receiving layer are used to set the active power impact of the cloud cover at the near moment as a parameter and introduce it into the prediction model. Assume that the number of neurons in the input layer, hidden layer, and output layer in the Elman neural network structure are M, N, and L respectively, and the connection weight matrices from the input layer to the hidden layer, from the hidden layer to the output layer, and from the receiving layer to the hidden layer are W respectively. 1ij (t), W 2jq (t), W 3rj (t). The nonlinear vector functions composed of the activation functions of the hidden layer units and the output layer units are f(g) and g(g) respectively.

[0120] Assume that the output of the hidden layer of the Elman neural network is:

[0121]

[0122] X Cr (t) = X j (t)

[0123] Among them, X L is the output value of the hidden layer; W 1ij (t), W 3rj(t) are the connection weight matrices from the input layer to the hidden layer and from the receiving layer to the hidden layer; U i (t) is the input vector of the hidden layer at the current moment; X Cr (t) is the vector of the receiving layer, which is used to remember the output value of the previous moment; X j (t) is the output value at the previous moment; f(g) is the nonlinear vector function composed of the excitation functions of the hidden layer units.

[0124] The output of the output layer is:

[0125]

[0126] Among them, y q (t) is the output value of the output layer; g(g) is the nonlinear vector function composed of the activation functions of the hidden layer units; W 2jq (t) is the connection weight matrix from the hidden layer to the output layer; X j (t) is the output value at the previous moment.

[0127] Step 209: Obtain basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located.

[0128] In an embodiment of the present invention, the basic information includes the location latitude α of the distributed photovoltaic power generation system; the target satellite remote sensing information includes the target cloud cover within a specified time period, and the target meteorological monitoring data includes the target temperature monitoring data, target total solar radiation monitoring data, target wind speed monitoring data and target relative humidity monitoring data within the specified time period.

[0129] It is worth noting that the specified time period can be set according to actual needs, and the embodiment of the present invention does not limit this. As an optional solution, the specified time period is 0-6 hours before the period to be predicted.

[0130] Step 210: Generate a distributed photovoltaic power change amount according to the target satellite remote sensing information through the pre-constructed cloud condition power change relationship.

[0131] In the embodiment of the present invention, the target satellite remote sensing information includes the target cloud cover within a specified time period.

[0132] Specifically, the target cloud amount within the specified time period is input into the cloud condition power change relationship for calculation, and the distributed photovoltaic power change within the specified time period is output.

[0133] Step 211: Generate the solar declination, hour angle and the difference in distance between the sun and the earth according to the time period to be predicted.

[0134] In the embodiment of the present invention, the period to be predicted is the time period for which the photovoltaic power needs to be predicted, for example, the period to be predicted is from the current time to 0-4 hours in the future. The start time of the period to be predicted is the current time.

[0135] Specifically, through Calculate the current time to generate the solar declination, where N is the day of the year and θ is the solar declination.

[0136] Specifically, the current time is calculated by t=15°×(T-12) to generate the solar declination, where T is the local standard time (hours); and t is the hour angle.

[0137] Specifically, the current time is calculated using the following formula to generate the difference in distance between the sun and the earth:

[0138]

[0139] θ0=(2πN-1) / 365

[0140] Among them, θ0 is the latitude of the sun's direct point, N is the day of the year, is the difference in distance between the Earth and the Sun.

[0141] Step 212: Generate a cosine value of the solar zenith angle according to the latitude of the distributed photovoltaic power generation system, the solar declination and the hour angle.

[0142] Specifically, by using sin h = sin α sin θ + sin α cos θ cost, the latitude, solar declination and hour angle of the distributed photovoltaic power generation system are calculated to obtain the sine value of the solar altitude angle sinh; the sine value of the solar altitude angle is calculated by the inverse sine function to generate the solar altitude angle h; and the cos θ cost is used to calculate the sine value of the solar altitude angle h. sun =cos h, calculate the solar altitude angle and generate the cosine value of the solar zenith angle.

[0143] Where h is the solar altitude angle; θ is the solar declination; t is the hour angle; α is the latitude of the distributed photovoltaic power generation system; cosθ sun is the cosine of the solar zenith angle.

[0144] Step 213: Generate the solar shortwave radiation at the upper boundary of the atmosphere according to the pre-acquired solar constant, the cosine value of the solar zenith angle, the latitude of the location of the distributed photovoltaic power generation system, the solar declination, the hour angle and the difference in the distance between the sun and the earth.

[0145] Specifically, through The solar constant, the cosine value of the solar zenith angle, the latitude of the distributed photovoltaic power generation system, the solar declination, the hour angle and the difference in the distance between the sun and the earth are calculated to generate the solar shortwave radiation. Among them, S is the solar shortwave radiation; S0 is the solar constant, and S0 is 1368W / m 2 ; θ is the solar declination; t is the hour angle; α is the latitude of the distributed photovoltaic power generation system; is the average distance between the sun and the earth; d is the distance between the sun and the earth; cosθ sun is the cosine of the solar zenith angle.

[0146] Step 214: Generate the distributed photovoltaic theoretical power for the time period to be predicted based on the solar energy resource utilization rate equation and the solar shortwave radiation at the upper boundary of the atmosphere.

[0147] Specifically, the solar shortwave radiation at the upper boundary of the atmosphere is input into the solar energy resource utilization rate relationship for calculation, and the distributed photovoltaic theoretical power for the predicted period is output.

[0148] Step 215: Generate the distributed photovoltaic predicted power for the time period to be predicted through the ultra-short-term power prediction model according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted.

[0149] In the embodiment of the present invention, the time period to be predicted is a time period during which photovoltaic power needs to be predicted, and the start time of the time period to be predicted is the current time.

[0150] Specifically, the target meteorological monitoring data, target satellite remote sensing information, distributed photovoltaic power change and distributed photovoltaic theoretical power in the time period to be predicted in the specified time period are input into the ultra-short-term power prediction model for prediction, and the distributed photovoltaic predicted power in the time period to be predicted is output. For example, if the time period to be predicted is 0-4 hours from the current time to the future, and the specified time period is 0-6 hours before the current time, then the target temperature monitoring data actually measured in 0-6 hours before the current time, the target total solar radiation monitoring data actually measured in 0-6 hours before the current time, the target wind speed monitoring data actually measured in 0-6 hours before the current time, the target relative humidity monitoring data actually measured in 0-6 hours before the current time, the target cloud cover in 0-6 hours before the current time, the distributed photovoltaic power change in 0-6 hours before the current time and the distributed photovoltaic theoretical power in the time period of the current time are input into the ultra-short-term power prediction model for photovoltaic power prediction, and the distributed photovoltaic predicted power in the future 0-4 hours is output.

[0151] The present invention combines the ultra-short-term power prediction of satellite remote sensing and meteorological monitoring data, can improve the traditional power prediction algorithm framework, and use effective data to improve the accuracy of regional distributed photovoltaic power prediction.

[0152] It is worth noting that the acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. The user information in the embodiments of this application is obtained through legal and compliant channels, and the acquisition, storage, use, and processing of user information are authorized and agreed by the customer.

[0153] It is worth noting that the information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of relevant countries and regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0154] It is worth noting that the technical solution provided in this application provides users with corresponding operation entrances for them to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered.

[0155] In the technical solution of the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information provided by the embodiment of the present invention, basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located are obtained; the distributed photovoltaic power change amount is generated according to the target satellite remote sensing information through the pre-constructed cloud condition power change relationship; the distributed photovoltaic theoretical power of the time period to be predicted is generated according to the time period to be predicted and the basic information through the pre-constructed solar energy resource utilization relationship; the distributed photovoltaic predicted power of the time period to be predicted is generated according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change amount and the distributed photovoltaic theoretical power of the time period to be predicted through the pre-constructed ultra-short-term power prediction model, and the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network, and combines satellite remote sensing information with real-time meteorological monitoring data of an automatic weather station to realize ultra-short-term power prediction of a large number of distributed photovoltaic power generation clusters within a large spatial range, based on solving the power generation grid-connected scheduling problem of new energy such as photovoltaics, maximizing the utilization efficiency of green energy, ensuring the safety of the power grid, and providing certain technical support for the safe and stable operation of the distribution network under small load and large power supply scenarios.

[0156] Figure 3 A schematic diagram of the structure of a distributed photovoltaic ultra-short-term power prediction device based on remote sensing information provided by an embodiment of the present invention, the device is used to execute the above-mentioned distributed photovoltaic ultra-short-term power prediction method based on remote sensing information, such as Figure 3 As shown, the device includes: a prediction data acquisition unit 11, a power change amount generation unit 12, a theoretical power generation unit 13 and an ultra-short-term power prediction unit 14.

[0157] The prediction data acquisition unit 11 is used to acquire basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located.

[0158] The power variation generating unit 12 is used to generate the distributed photovoltaic power variation according to the target satellite remote sensing information through the pre-constructed cloud condition power variation relationship.

[0159] The theoretical power generation unit 13 is used to generate the distributed photovoltaic theoretical power of the time period to be predicted according to the time period to be predicted and the basic information through the pre-constructed solar energy resource utilization relationship.

[0160] The ultra-short-term power prediction unit 14 is used to generate the distributed photovoltaic predicted power for the time period to be predicted through a pre-constructed ultra-short-term power prediction model according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted. The ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network.

[0161] In the embodiment of the present invention, the device further includes: a historical data acquisition unit 15 , a power time series data generation unit 16 , a synchronous time series data generation unit 17 and a relational expression construction unit 18 .

[0162] The historical data acquisition unit 15 is used to acquire the historical active power and operation status data of the distributed photovoltaic power generation system, the historical satellite remote sensing information and the historical meteorological monitoring data in the area where the distributed photovoltaic power generation system is located.

[0163] The power time series data generating unit 16 is used to generate distributed photovoltaic historical power time series data according to the preset time resolution, based on the basic information, the operating status data and the historical active power.

[0164] The synchronous time series data generating unit 17 is used to synchronize the historical satellite remote sensing information and the historical meteorological monitoring data according to the distributed photovoltaic historical power time series data, and generate the historical satellite remote sensing time series data and the historical meteorological monitoring time series data.

[0165] The relational expression constructing unit 18 is used to construct a cloud condition power change relational expression and a solar energy resource utilization relational expression based on basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data.

[0166] In an embodiment of the present invention, the relationship construction unit 18 is specifically used to extract effective data from the distributed photovoltaic historical power time series data according to the basic information, and determine the distributed photovoltaic historical effective power data; according to the target time period corresponding to the distributed photovoltaic historical effective power data, extract the historical satellite remote sensing data and the historical meteorological monitoring data from the historical satellite remote sensing time series data and the historical meteorological monitoring time series data; construct a cloud condition power change relationship based on the historical satellite remote sensing data and the distributed photovoltaic historical effective power data; construct a solar energy resource utilization relationship based on the historical satellite remote sensing data, the historical meteorological monitoring data and the distributed photovoltaic historical effective power data.

[0167] In an embodiment of the present invention, the historical contemporaneous satellite remote sensing data includes the historical contemporaneous cloud cover; the relationship constructing unit 18 is specifically used to determine the non-zero cloud cover period according to the historical contemporaneous cloud cover; the historical contemporaneous cloud cover corresponding to the non-zero cloud cover period is screened out from the historical contemporaneous cloud cover, and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud cover period is screened out from the distributed photovoltaic historical effective power data; according to the preset cloud cover level mapping relationship, the historical contemporaneous cloud cover corresponding to the non-zero cloud cover period is graded to generate the cloud cover level corresponding to each non-zero cloud cover period; according to the cloud cover level corresponding to each non-zero cloud cover period and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud cover period, a cloud condition power linear regression model is constructed; the cloud condition power linear regression model is parameterized by the least squares method to generate a cloud condition power change relationship.

[0168] In an embodiment of the present invention, the historical satellite remote sensing data for the same period includes the historical cloud cover for the same period, and the historical meteorological monitoring data for the same period includes the historical total solar radiation data for the same period; the relationship construction unit 18 is specifically used to determine the zero cloud period according to the historical cloud cover for the same period; filter out the total solar radiation data corresponding to the zero cloud period from the historical total solar radiation data, and filter out the distributed photovoltaic historical effective power data corresponding to the zero cloud period from the distributed photovoltaic historical effective power data; construct a solar energy resource utilization linear regression model based on the total solar radiation data corresponding to the zero cloud period and the distributed photovoltaic historical effective power data corresponding to the zero cloud period; solve the parameters of the solar energy resource utilization linear regression model through the least squares method to generate a solar energy resource utilization relationship.

[0169] In an embodiment of the present invention, the basic information includes the location latitude of the distributed photovoltaic power generation system; according to the time period to be predicted, the solar declination, hour angle and the difference in distance between the sun and the earth are generated; the theoretical power generation unit 13 is specifically used to generate the cosine value of the solar zenith angle according to the location latitude of the distributed photovoltaic power generation system, the solar declination and hour angle; according to the pre-acquired solar constant, the cosine value of the solar zenith angle, the location latitude of the distributed photovoltaic power generation system, the solar declination, hour angle and the difference in distance between the sun and the earth, the solar shortwave radiation at the upper boundary of the atmosphere is generated; through the solar energy resource utilization relationship, the distributed photovoltaic theoretical power of the time period to be predicted is generated according to the solar shortwave radiation at the upper boundary of the atmosphere.

[0170] In the scheme of the embodiment of the present invention, basic information of the distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located are obtained; a distributed photovoltaic power change is generated according to the target satellite remote sensing information through a pre-constructed cloud condition power change relationship; a distributed photovoltaic theoretical power of the time period to be predicted is generated according to the time period to be predicted and the basic information through a pre-constructed solar energy resource utilization relationship; a distributed photovoltaic predicted power of the time period to be predicted is generated according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power of the time period to be predicted through a pre-constructed ultra-short-term power prediction model, the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network, and combines satellite remote sensing information with real-time meteorological monitoring data of an automatic weather station to realize ultra-short-term power prediction of a large number of distributed photovoltaic power generation clusters within a large spatial range, based on solving the problem of power generation and grid-connected scheduling of new energy such as photovoltaics, maximizing the utilization efficiency of green energy, ensuring the safety of the power grid, and providing certain technical support for the safe and stable operation of the distribution network under small load and large power supply scenarios.

[0171] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer device, and specifically, the computer device may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0172] An embodiment of the present invention provides a computer device, including a memory and a processor, the memory is used to store information including program instructions, the processor is used to control the execution of the program instructions, and when the program instructions are loaded and executed by the processor, the steps of the embodiment of the above-mentioned distributed photovoltaic ultra-short-term power prediction method based on remote sensing information are implemented. For a specific description, please refer to the embodiment of the above-mentioned distributed photovoltaic ultra-short-term power prediction method based on remote sensing information.

[0173] Reference below Figure 4 , which shows a schematic diagram of the structure of a computer device 600 suitable for implementing an embodiment of the present application.

[0174] like Figure 4 As shown, the computer device 600 includes a central processing unit (CPU) 601, which can perform various appropriate operations and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the computer device 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0175] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal feedback device (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed as needed as the storage section 608.

[0176] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 609, and / or installed from the removable medium 611.

[0177] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0178] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0179] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0180] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0181] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0182] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0183] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0184] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0185] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0186] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0187] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0188] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A distributed photovoltaic ultra-short-term power prediction method based on remote sensing information, characterized in that: The method comprises: Obtaining basic information of a distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located; Generate distributed photovoltaic power variation according to the target satellite remote sensing information through a pre-constructed cloud condition power variation relationship; Generate the distributed photovoltaic theoretical power of the time period to be predicted according to the time period to be predicted and the basic information through the pre-constructed solar energy resource utilization rate relationship; Through a pre-constructed ultra-short-term power prediction model, the distributed photovoltaic predicted power for the time period to be predicted is generated according to the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted. The ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network.

2. The distributed photovoltaic ultra-short-term power prediction method based on remote sensing information according to claim 1 is characterized in that: The method further comprises: Acquire historical active power and operating status data of the distributed photovoltaic power generation system, historical satellite remote sensing information and historical meteorological monitoring data in the area where the distributed photovoltaic power generation system is located; Generate distributed photovoltaic historical power time series data according to the basic information, operating status data and historical active power at a preset time resolution; According to the distributed photovoltaic historical power time series data, the historical satellite remote sensing information and the historical meteorological monitoring data are synchronized to generate the historical satellite remote sensing time series data and the historical meteorological monitoring time series data; The cloud condition power change relationship and the solar energy resource utilization relationship are constructed based on the basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data.

3. The distributed photovoltaic ultra-short-term power prediction method based on remote sensing information according to claim 2 is characterized in that: The cloud condition power change relationship and solar energy resource utilization relationship are constructed based on the basic information, distributed photovoltaic historical power time series data, historical satellite remote sensing time series data and historical meteorological monitoring time series data, including: Extracting effective data from the distributed photovoltaic historical power time series data according to the basic information to determine the distributed photovoltaic historical effective power data; According to the target period corresponding to the distributed photovoltaic historical effective power data, extracting the historical satellite remote sensing data and the historical meteorological monitoring data for the same period from the historical satellite remote sensing time series data and the historical meteorological monitoring time series data; Constructing the cloud condition power change relationship according to the historical satellite remote sensing data and the distributed photovoltaic historical effective power data for the same period; A solar energy resource utilization rate relationship is constructed based on the historical satellite remote sensing data, the historical meteorological monitoring data and the distributed photovoltaic historical effective power data for the same period.

4. The distributed photovoltaic ultra-short-term power prediction method based on remote sensing information according to claim 3 is characterized in that: The satellite remote sensing data for the same period in history include cloud cover for the same period in history; The cloud condition power change relationship is constructed based on the historical satellite remote sensing data and the distributed photovoltaic historical effective power data for the same period, including: Determine a non-zero cloud cover period according to the historical cloud cover for the same period; Filter out the historical cloud amount corresponding to the non-zero cloud amount period from the historical cloud amount in the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the non-zero cloud amount period from the distributed photovoltaic historical effective power data; According to the preset cloud amount level mapping relationship, the historical cloud amounts corresponding to the non-zero cloud amount period are classified into levels, and the cloud amount level corresponding to each non-zero cloud amount period is generated; Constructing a cloud condition power linear regression model according to the cloud level corresponding to each non-zero cloud period and the distributed photovoltaic historical effective power data corresponding to the non-zero cloud period; The cloud condition power linear regression model is parameterized by the least squares method to generate a cloud condition power variation relationship.

5. The distributed photovoltaic ultra-short-term power prediction method based on remote sensing information according to claim 3 is characterized in that: The satellite remote sensing data for the same period in history include cloud cover for the same period in history, and the meteorological monitoring data for the same period in history include total solar radiation data for the same period in history; The solar energy resource utilization rate relationship is constructed based on the historical satellite remote sensing data, the historical meteorological monitoring data and the distributed photovoltaic historical effective power data, including: Determine a zero cloud period based on the historical cloud cover for the same period; Filter out the total solar radiation data corresponding to the zero cloud period from the historical total solar radiation data for the same period, and filter out the distributed photovoltaic historical effective power data corresponding to the zero cloud period from the distributed photovoltaic historical effective power data; Constructing a solar energy resource utilization linear regression model based on the total solar radiation data corresponding to the zero cloud period and the distributed photovoltaic historical effective power data corresponding to the zero cloud period; The solar energy resource utilization rate linear regression model is parameter-solved by the least square method to generate the solar energy resource utilization rate relationship.

6. The distributed photovoltaic ultra-short-term power prediction method based on remote sensing information according to claim 1 is characterized in that: The basic information includes the location latitude of the distributed photovoltaic power generation system; The method generates the distributed photovoltaic theoretical power of the time period to be predicted according to the time period to be predicted and the basic information through the pre-constructed solar energy resource utilization rate relationship, including: According to the time period to be predicted, generating solar declination, hour angle and difference in distance between the sun and the earth; Generate a cosine value of the solar zenith angle according to the latitude of the location of the distributed photovoltaic power generation system, the solar declination and the hour angle; Generate solar shortwave radiation at the upper boundary of the atmosphere based on the pre-acquired solar constant, the cosine value of the solar zenith angle, the latitude of the location of the distributed photovoltaic power generation system, the solar declination, the hour angle and the difference in the distance between the sun and the earth; The distributed photovoltaic theoretical power for the time period to be predicted is generated through the solar energy resource utilization rate relationship according to the solar shortwave radiation at the upper boundary of the atmosphere.

7. A distributed photovoltaic ultra-short-term power prediction device based on remote sensing information, characterized in that: The device comprises: A prediction data acquisition unit, used to acquire basic information of a distributed photovoltaic power generation system within a specified time period and target satellite remote sensing information and target meteorological monitoring data in the area where the distributed photovoltaic power generation system is located; A power variation generating unit, used to generate a distributed photovoltaic power variation according to the target satellite remote sensing information by using a pre-constructed cloud condition power variation relationship; A theoretical power generation unit, used to generate a distributed photovoltaic theoretical power for a time period to be predicted according to a pre-built solar energy resource utilization rate relationship, the time period to be predicted and the basic information; The ultra-short-term power prediction unit is used to generate the distributed photovoltaic predicted power for the time period to be predicted based on the target meteorological monitoring data, the target satellite remote sensing information, the distributed photovoltaic power change and the distributed photovoltaic theoretical power for the time period to be predicted through a pre-built ultra-short-term power prediction model, wherein the ultra-short-term power prediction model is obtained by training based on a dynamic recursive neural network.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information as described in any one of claims 1 to 6 is implemented.

9. A computer device comprising a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, characterized in that: When the program instructions are loaded and executed by the processor, the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the distributed photovoltaic ultra-short-term power prediction method based on remote sensing information as described in any one of claims 1 to 6 is implemented.

Images