A photovoltaic power station shadow loss electric quantity prediction method and system based on big data

Abstract

The application discloses a photovoltaic power station shadow loss electric quantity prediction method and system based on big data. The unit installed power curve of each inverter of a to-be-tested power station in a prediction period is determined; a plurality of unit installed power curves are fitted to obtain a benchmark inverter unit installed power curve; shadow identification is performed on each inverter to respectively determine a shadow period of the corresponding inverter; the area between the unit installed power curve of the corresponding inverter in the shadow period and the benchmark inverter unit installed power curve is calculated to obtain the loss electric quantity of the corresponding inverter in the shadow period; and the loss electric quantity sum of each inverter in the prediction period is calculated to obtain the shadow loss electric quantity of the to-be-tested power station in the prediction period. The application can reduce data processing complexity, improve prediction efficiency, is not dependent on a large amount of meteorological data, can predict the loss electric quantity based on the shadow period of each inverter, and can improve the overall accuracy of predicting the loss electric quantity of the power station.

Description

Technical Field

[0001] This invention relates to the field of solar power generation technology, and in particular to a method and system for predicting the power loss due to shading in photovoltaic power plants based on big data. Background Technology

[0002] Applying artificial intelligence and big data technologies to the operation and management of photovoltaic power generation projects can help analyze whether a power station is experiencing shading by analyzing trends in weather conditions and power generation, as well as massive datasets from thousands of solar power plants. Current methods for predicting shading loss in photovoltaic power plants rely on nearly a hundred types of astronomical data, including Earth's heliocentric longitude and latitude, Sun's geocentric longitude and latitude, Earth's average anomaly angle, Moon's average anomaly angle, Moon's ascending node distance, true ecliptic oblique angle, average ecliptic oblique angle, and geocentric solar right ascension declination. This data processing is highly complex and difficult to implement, and the inclusion of numerous multi-level approximations in the data further impacts prediction accuracy. Summary of the Invention

[0003] To address the aforementioned problems, the inventors have developed this invention, which, through specific implementation methods, provides a method and system for predicting shading loss power generation in photovoltaic power plants based on big data.

[0004] In a first aspect, embodiments of the present invention provide a method for predicting shading loss power generation in photovoltaic power plants based on big data, comprising the following steps:

[0005] Divide the AC output power of each inverter in the power station under test within the prediction period by the capacity of the corresponding inverter to obtain the unit installed power curve of the corresponding inverter.

[0006] The unit installed power curves of multiple inverters are fitted to obtain the unit installed power curve of the benchmark inverter.

[0007] Shadow identification is performed for each inverter, and the shadow period of the corresponding inverter is determined.

[0008] Calculate the area between the unit installed power curve of the inverter corresponding to the shaded period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shaded period;

[0009] Calculate the total power loss of each inverter during the prediction period to obtain the power loss due to shadowing of the power station under test during the prediction period.

[0010] Specifically, the unit installed power curves of multiple inverters are fitted to obtain the benchmark inverter unit installed power curve, including the following steps:

[0011] Determine the average value of the unit installed power curves of multiple inverters at the same time, obtain the corresponding value of the unit installed power curve of the benchmark inverter at that time, and plot the unit installed power curve of the benchmark inverter based on the time and the corresponding value of the unit installed power curve of the benchmark inverter at different times.

[0012] Specifically, shadow identification is performed on each inverter to determine the shadow period for each inverter, including the following steps:

[0013] The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period.

[0014] Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day.

[0015] Based on the sunny day identification parameters, sunny days within the prediction period are determined, and all non-sunny days are defined as the shadow period for all inverters.

[0016] Determine the hourly average DC current dispersion rate of each inverter during sunny days within the forecast period;

[0017] When the hourly average DC current dispersion rate exceeds a preset threshold, the corresponding hour is determined as the shadow period of the corresponding inverter.

[0018] Determine the total shadow periods for each inverter within the prediction period.

[0019] Specifically, based on the daily declination angle, sunset angle, and latitude of the power station under test within the prediction period, the atmospheric surface radiation received by the power station under test on the corresponding day is determined, including the following steps:

[0020] The daily declination angle within the prediction period is determined by the following formula:

[0021] sigma i =ab*cos(θ)+c*sin(θ)-d*cos(2θ)+e*sin(2θ)-f*cos(3θ)+g*sin(3θ),

[0022]

[0023] Among them, sigma i θ represents the solar declination angle on the i-th day. The values ​​of parameters a, b, c, d, e, f, and g are derived from the Spencer operator. i represents the natural day number in the natural year in which the prediction period is located, and θ represents the solar angle.

[0024] The sunset hour angle for each day within the prediction period is determined by the following formula:

[0025] Among them omega i Let represent the sunset angle on the i-th day, lat represent the latitude of the power station to be measured, pi represent pi, and acos represent the inverse cosine function;

[0026] The atmospheric surface radiation received by the power station under test on day i within the prediction period is determined by the following formula.

[0027] Among them, TOA i Let A represent the amount of atmospheric surface radiation received by the power station under test on day i, and let A represent the solar constant. ifit This represents the correction value caused by the distance between the sun and the earth on the i-th day of the test power station.

[0028] Specifically, determining sunny days within the prediction period based on the aforementioned sunny day identification parameters includes the following steps:

[0029] The sunny day identification parameters and sunny day thresholds for each day within the prediction period are compared, and the natural days whose sunny day identification parameters exceed the sunny day thresholds are determined as sunny days.

[0030] Specifically, determining the hourly average DC current dispersion rate for each inverter during sunny days within the forecast period includes the following steps:

[0031] The hourly average DC current dispersion rate for each inverter during sunny days within the prediction period is determined by the following formula.

[0032]

[0033] Among them, dc dis represents the hourly average DC current dispersion rate of the corresponding inverter during a sunny day, and std represents the standard deviation of the DC current of each string corresponding to the inverter. This represents the average DC current of each string corresponding to the inverter, where the hourly average DC current value is the average of all instantaneous DC currents within the corresponding hour.

[0034] Secondly, embodiments of the present invention provide a photovoltaic power plant shading loss power prediction system based on big data, comprising:

[0035] The curve plotting module is used to divide the AC output power of each inverter in the power station under test by the corresponding inverter capacity during the prediction period to obtain the unit installed power curve of the corresponding inverter; and to fit the unit installed power curves of multiple inverters to obtain the benchmark inverter unit installed power curve.

[0036] The shadow recognition module is used to identify shadows for each inverter and determine the shadow period for each inverter.

[0037] The power loss prediction module is used to calculate the area between the unit installed power curve of the inverter corresponding to the shadow period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shadow period; and to calculate the total power loss of each inverter in the prediction period to obtain the shadow power loss of the power station under test in the prediction period.

[0038] Thirdly, embodiments of the present invention provide a method for identifying shadows in photovoltaic power plants based on big data, comprising the following steps:

[0039] The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period.

[0040] Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day.

[0041] Based on the sunny day identification parameters, sunny days within the prediction period are determined, and the entire 24 hours of non-sunny days are defined as the shadow period for all inverters of the power station.

[0042] Specifically, the big data-based photovoltaic power plant shadow identification method further includes the following steps:

[0043] Determine the hourly average DC current dispersion rate of each inverter in the power station during sunny days within the prediction period;

[0044] When the hourly average DC current dispersion rate exceeds a preset threshold, the corresponding hour is determined as the shadow period of the corresponding inverter.

[0045] Based on the same inventive concept, embodiments of the present invention also provide a computer storage medium storing instructions, which, when executed, implement the aforementioned method for predicting power loss due to shading in photovoltaic power plants based on big data or the method for identifying shading in photovoltaic power plants based on big data.

[0046] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0047] The big data-based photovoltaic power plant shading loss prediction scheme proposed in this invention does not rely on a large amount of meteorological data, which can reduce data processing complexity and improve prediction efficiency. Based on the shading period of each inverter, the power loss prediction scheme can improve the overall accuracy of power plant power loss prediction.

[0048] The big data-based photovoltaic power plant shadow identification scheme proposed in this invention does not rely on a large amount of meteorological data, reduces data processing complexity, is easy to implement, and can achieve rapid and accurate identification. Furthermore, it can achieve hourly shadow identification with high prediction accuracy and high precision, facilitating the application of shadow identification data.

[0049] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0050] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0052] Figure 1 This is a flowchart of a method for predicting the shading loss power of a photovoltaic power plant based on big data, as described in an embodiment of the present invention.

[0053] Figure 2 This is a graph showing the unit installed power in an embodiment of the present invention;

[0054] Figure 3 This is a block diagram of a photovoltaic power plant shadow loss power prediction system based on big data, as described in an embodiment of the present invention.

[0055] Figure 4 This is a flowchart of a photovoltaic power station shadow recognition method based on big data in an embodiment of the present invention. Detailed Implementation

[0056] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0057] To address the problems existing in the prior art, embodiments of the present invention provide a method and system for predicting power loss due to shading in photovoltaic power plants based on big data, and a method and system for identifying shading in photovoltaic power plants based on big data.

[0058] This invention provides a method for predicting shading loss power generation in photovoltaic power plants based on big data, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0059] Step S1: Divide the AC output power of each inverter in the power station under test within the prediction period by the corresponding inverter capacity to obtain the unit installed power curve of the corresponding inverter; fit the unit installed power curves of multiple inverters to obtain the benchmark inverter unit installed power curve.

[0060] In some specific embodiments, the unit installed power curves of multiple inverters are fitted to obtain the unit installed power curve of a benchmark inverter. This includes the following steps: determining the average value of the unit installed power curves of multiple inverters at the same time, obtaining the corresponding value of the unit installed power curve of the benchmark inverter at the same time, and plotting the unit installed power curve of the benchmark inverter based on the time and the corresponding value of the unit installed power curve of the benchmark inverter at different times.

[0061] In some specific embodiments, the median curve of the power curves of all inverters under the power station is obtained as the benchmark inverter equipment power curve under the power station, with the default benchmark equipment installed capacity being 1kW.

[0062] Step S2: Perform shadow identification for each inverter and determine the shadow period for the corresponding inverter.

[0063] In some specific embodiments, shadow identification is performed on each inverter to determine the shadow period for each inverter. This includes the following steps: determining the atmospheric surface radiation received by the power station on the corresponding day based on the declination angle, sunset angle, and latitude of the power station under test for each day within the prediction period; dividing the atmospheric surface radiation received by the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear day identification parameter for the corresponding day; determining clear days within the prediction period based on the clear day identification parameter, and defining all non-clear days as the shadow period for all inverters; determining the hourly average DC current dispersion rate of each inverter on clear days within the prediction period; when the hourly average DC current dispersion rate exceeds a preset threshold, determining the corresponding hour as the shadow period for the corresponding inverter; and determining all shadow periods for each inverter within the prediction period.

[0064] In some specific embodiments, the atmospheric surface radiation received by the power station on a given day is determined based on the declination angle, sunset angle, and latitude of the power station to be measured each day within the prediction period, including the following steps:

[0065] The daily declination angle within the prediction period is determined by the following formula:

[0066] sigma i =ab*cos(θ)+c*sin(θ)-d*cos(2θ)+e*sin(2θ)-f*cos(3θ)+g*sin(3θ),

[0067]

[0068] Among them, sigma i θ represents the solar declination angle on the i-th day. The values ​​of parameters a, b, c, d, e, f, and g are derived from the Spencer operator. i represents the natural day number in the natural year in which the prediction period is located, and θ represents the solar angle.

[0069] The Spencer operator was proposed by Spencer, where the solar declination angle is in radians. The algorithm formula is as follows:

[0070] sigma i =0.006918-0.399912cos(θ)+0.070257sin(θ)-0.006758cos(2θ)+0.000907sin(2θ)-0.002697cos(3θ)+0.00148sin(3θ)

[0071] The sunset hour angle for each day within the prediction period is determined by the following formula:

[0072] Among them omega i Let represent the sunset angle on the i-th day, lat represent the latitude of the power station to be measured, pi represent pi, and acos represent the inverse cosine function;

[0073] The atmospheric surface radiation received by the power station under test on day i within the prediction period is determined by the following formula.

[0074] Among them, TOA i Let A represent the amount of atmospheric surface radiation received by the power station under test on day i, and let A represent the solar constant. ifit This represents the correction value caused by the Earth-Sun distance on day i of the test power station. The solar constant represents the intensity of solar radiation reaching the upper boundary of the atmosphere vertically. In 1981, the World Meteorological Organization (WMO) published the solar constant value as 1367 ± 7 W / m². 2 .

[0075] via dis ifit =h + j*cos(i) + k*sin(i) + l*cos(2*i) + m*sin(2*i), determine dis on day i. ifitThe values ​​of the parameters h, j, k, l, and m are derived from the Iqbal (1983) model, a classic meteorological model of solar radiation, first documented in "An introduction to solar radiation" published in 1983 by Muhammad Iqbal, a scholar at the University of British Columbia, Canada.

[0076] In some specific embodiments, the amount of radiation received at the atmospheric surface of the power station under test is divided by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day, specifically including:

[0077] KT i =TOA i / GHI i

[0078] Among them, KT i TOA represents the clear weather identification parameter for day i in the prediction period of the power station under test. i GHI represents the amount of atmospheric surface radiation received by the power station under test on day i. i GHI represents the total horizontal radiation of the power station on day i. i Meteorological data originating from weather stations.

[0079] In some specific embodiments, determining sunny days within a prediction period based on the sunny day identification parameters includes the following steps: comparing the sunny day identification parameters for each day within the prediction period with a sunny day threshold, and determining the natural days whose sunny day identification parameters exceed the sunny day threshold as sunny days. For example, comparing the sunny day identification parameters for each day within the prediction period with a sunny day threshold of 0.6, and determining the natural days whose sunny day identification parameters exceed the sunny day threshold as sunny days.

[0080] In some specific embodiments, the entire 24 hours of a non-sunny day are defined as the shadow period for all inverters; shadow periods also exist on sunny days, therefore, shadow identification is further performed for each hour of a sunny day.

[0081] In some specific embodiments, determining the hourly average DC current dispersion rate for each inverter during sunny days within the prediction period includes the following steps: determining the hourly average DC current dispersion rate for each inverter during sunny days within the prediction period using the following formulas. Among them, dc dis This represents the hourly average DC current dispersion rate of the corresponding inverter during a sunny day, and std represents the standard deviation of the DC current of each string corresponding to the inverter. This represents the average DC current of each string corresponding to the inverter, where the hourly average DC current value is the average of all instantaneous DC currents within the corresponding hour.

[0082] One inverter corresponds to multiple strings, and each string includes multiple photovoltaic (PV) modules. For example, if the average DC current dispersion rate of multiple strings corresponding to one inverter is greater than a preset threshold of 0.2 within the same hour on a sunny day, it is determined that there is a shadow within that hour, and the corresponding hour is the shadow period. The average DC current dispersion rate of each inverter per hour on a sunny day is compared with 0.2 to determine the shadow period of each inverter on a sunny day.

[0083] Step S3: Calculate the area between the unit installed power curve of the inverter corresponding to the shaded period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shaded period; calculate the total power loss of each inverter in the prediction period to obtain the power loss of the power station under test during the shaded period.

[0084] In a specific embodiment, the unit installed power curve is as follows: Figure 2 As shown, the horizontal axis represents time, and the vertical axis represents the AC output power of the inverter divided by the installed capacity of the same inverter. Figure 2 In the diagram, the lower original equipment power curve represents the unit installed power curve of the inverter during the shaded period, while the upper benchmark equipment power curve represents the unit installed power curve of the benchmark inverter. Specifically, based on the start and end coordinates of the shaded period, the area between the unit installed power curve of the inverter corresponding to the shaded period and the unit installed power curve of the benchmark inverter is obtained through integration. This area represents the power loss of the corresponding inverter during the shaded period. If the one-hour period between 8 and 9 o'clock is the shaded period, then the area between the original equipment power curve and the benchmark equipment power curve between 8 and 9 o'clock is calculated to obtain the power loss of the inverter corresponding to 8 and 9 o'clock.

[0085] In some specific embodiments, the power curve of the benchmark device is overlapped or shifted with the power curve of the device with shadows to eliminate the gap or unclosed area between the two curves, and then the area between the two curves is calculated.

[0086] In the above method of this embodiment, the photovoltaic power plant shading loss power prediction scheme based on big data proposed in this invention does not rely on a large amount of meteorological data, can reduce data processing complexity, improve prediction efficiency, and predict power loss based on the shading period of each inverter, which can improve the overall accuracy of predicting power plant power loss.

[0087] Those skilled in the art can change the above order without departing from the scope of protection of this disclosure.

[0088] Another embodiment of the present invention provides a photovoltaic power plant shading loss power prediction system based on big data, such as... Figure 3 As shown, it includes:

[0089] The curve plotting module is used to divide the AC output power of each inverter in the power station under test by the corresponding inverter capacity during the prediction period to obtain the unit installed power curve of the corresponding inverter; and to fit the unit installed power curves of multiple inverters to obtain the benchmark inverter unit installed power curve.

[0090] The shadow recognition module is used to identify shadows for each inverter and determine the shadow period for each inverter.

[0091] The power loss prediction module is used to calculate the area between the unit installed power curve of the inverter corresponding to the shadow period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shadow period; and to calculate the total power loss of each inverter in the prediction period to obtain the shadow power loss of the power station under test in the prediction period.

[0092] Regarding the system in the above embodiments, the specific manner in which each module performs its operations has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0093] In this embodiment, the photovoltaic power plant shading loss prediction scheme based on big data proposed in this invention does not rely on a large amount of meteorological data, which can reduce data processing complexity and improve prediction efficiency. Based on the shading period of each inverter, the power loss prediction scheme can improve the overall accuracy of predicting power plant power loss.

[0094] Another embodiment of the present invention provides a method for identifying shadows in photovoltaic power plants based on big data, such as... Figure 4 As shown, it includes the following steps:

[0095] The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period.

[0096] Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day.

[0097] Based on the sunny day identification parameters, sunny days within the prediction period are determined, and the entire 24 hours of non-sunny days are defined as the shadow period for all inverters of the power station.

[0098] In some specific embodiments, the big data-based photovoltaic power plant shadow identification method further includes the following steps:

[0099] Determine the hourly average DC current dispersion rate of each inverter in the power station during sunny days within the prediction period; when the hourly average DC current dispersion rate exceeds a preset threshold, determine the corresponding hour as the shaded period of the corresponding inverter.

[0100] In the above embodiments, the photovoltaic power station shadow identification scheme based on big data proposed in this invention does not rely on a large amount of meteorological data, reduces the complexity of data processing, is easy to implement, and can achieve fast and accurate identification. Furthermore, it can achieve hourly shadow identification with high prediction accuracy and high precision, facilitating the application of shadow identification data.

[0101] Based on the same inventive concept, embodiments of the present invention also provide a computer storage medium storing instructions, which, when executed, implement the aforementioned method for predicting power loss due to shading in photovoltaic power plants based on big data or the method for identifying shading in photovoltaic power plants based on big data.

[0102] Any modifications, additions, and equivalent substitutions made within the scope of the principles of this invention shall still fall within the patent coverage of this invention.

Claims

1. A method for predicting shading loss power generation in photovoltaic power plants based on big data, characterized in that, Includes the following steps: Divide the AC output power of each inverter in the power station under test within the prediction period by the capacity of the corresponding inverter to obtain the unit installed power curve of the corresponding inverter. The unit installed power curves of multiple inverters are fitted to obtain the unit installed power curve of the benchmark inverter. Shadow identification is performed for each inverter, and the shadow period of the corresponding inverter is determined. Calculate the area between the unit installed power curve of the inverter corresponding to the shaded period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shaded period; Calculate the total power loss of each inverter during the prediction period to obtain the power loss due to shadowing of the power station under test during the prediction period; Shadow identification is performed for each inverter to determine the shadow period for each inverter, including the following steps: The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period. Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day. Based on the sunny day identification parameters, sunny days within the prediction period are determined, and all non-sunny days are defined as the shadow period for all inverters. Determine the hourly average DC current dispersion rate of each inverter during sunny days within the forecast period; When the hourly average DC current dispersion rate exceeds a preset threshold, the corresponding hour is determined as the shadow period of the corresponding inverter. Determine all shadow periods for each inverter within the prediction period; The atmospheric surface radiation received by the power station is determined based on the daily declination angle, sunset angle, and latitude of the power station under test within the prediction period, including the following steps: The daily declination angle within the prediction period is determined by the following formula: ， ， in, Indicates the first i The solar declination angle of the day, with parameters a, b, c, d, e, f, and g, are derived from the Spencer operator. i This indicates the ordinal number of the natural day within the calendar year in which the prediction period is located. Indicates the sun angle; The sunset hour angle for each day within the prediction period is determined by the following formula: ,in Indicates the first i At sunset, Indicates the latitude of the power station to be measured. Represents pi (π). Represents the inverse cosine function; The following formula is used to determine the number of power stations to be measured within the prediction period. i The amount of radiation received by the surface of the sun's atmosphere. ,in, Indicates the power station under test. i The amount of radiation received by the surface of the sun's atmosphere. A Represents the solar constant. Indicates the power station under test. i The correction value caused by the distance between the Sun and the Earth.

2. The method as described in claim 1, characterized in that, The unit installed power curves of multiple inverters are fitted to obtain the unit installed power curve of a benchmark inverter, including the following steps: Determine the average value of the unit installed power curves of multiple inverters at the same time, obtain the corresponding value of the unit installed power curve of the benchmark inverter at that time, and plot the unit installed power curve of the benchmark inverter based on the time and the corresponding value of the unit installed power curve of the benchmark inverter at different times.

3. The method as described in claim 1, characterized in that, Based on the sunny day identification parameters, the sunny days within the prediction period are determined, including the following steps: The sunny day identification parameters and sunny day thresholds for each day within the prediction period are compared, and the natural days whose sunny day identification parameters exceed the sunny day thresholds are determined as sunny days.

4. The method as described in claim 1, characterized in that, Determining the hourly average DC current dispersion rate for each inverter during sunny days within the forecast period includes the following steps: The hourly average DC current dispersion rate for each inverter during sunny days within the prediction period is determined by the following formula. in, This represents the hourly average DC current dispersion rate of the corresponding inverter during a sunny day. This represents the standard deviation of the DC current in each string corresponding to the inverter. This represents the average DC current of each string corresponding to the inverter, where the hourly average DC current value is the average of all instantaneous DC currents within the corresponding hour.

5. A photovoltaic power plant shading loss power prediction system based on big data, characterized in that, include: The curve plotting module is used to divide the AC output power of each inverter in the power station under test by the corresponding inverter capacity during the prediction period to obtain the unit installed power curve of the corresponding inverter; and to fit the unit installed power curves of multiple inverters to obtain the benchmark inverter unit installed power curve. The shadow recognition module is used to identify shadows for each inverter and determine the shadow period for each inverter. The power loss prediction module is used to calculate the area between the unit installed power curve of the inverter corresponding to the shadow period and the unit installed power curve of the benchmark inverter to obtain the power loss of the corresponding inverter during the shadow period; and to calculate the total power loss of each inverter in the prediction period to obtain the shadow power loss of the power station under test in the prediction period. Shadow identification is performed for each inverter to determine the shadow period for each inverter, including the following steps: The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period. Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day. Based on the sunny day identification parameters, sunny days within the prediction period are determined, and all non-sunny days are defined as the shadow period for all inverters. Determine the hourly average DC current dispersion rate of each inverter during sunny days within the forecast period; When the hourly average DC current dispersion rate exceeds a preset threshold, the corresponding hour is determined as the shadow period of the corresponding inverter. Determine all shadow periods for each inverter within the prediction period; The atmospheric surface radiation received by the power station is determined based on the daily declination angle, sunset angle, and latitude of the power station under test within the prediction period, including the following steps: The daily declination angle within the prediction period is determined by the following formula: ， ， in, Indicates the first i The solar declination angle of the day, with parameters a, b, c, d, e, f, and g, are derived from the Spencer operator. i This indicates the ordinal number of the natural day within the calendar year in which the prediction period is located. Indicates the sun angle; The sunset hour angle for each day within the prediction period is determined by the following formula: ,in Indicates the first i At sunset, Indicates the latitude of the power station to be measured. Represents pi (π). Represents the inverse cosine function; The following formula is used to determine the number of power stations to be measured within the prediction period. i The amount of radiation received by the surface of the sun's atmosphere. ,in, Indicates the power station under test. i The amount of radiation received by the surface of the sun's atmosphere. A Represents the solar constant. Indicates the power station under test. i The correction value caused by the distance between the Sun and the Earth.

6. A method for identifying shadows in photovoltaic power plants based on big data, characterized in that, Includes the following steps: The amount of atmospheric surface radiation received by the power station under test on the corresponding day is determined based on the declination angle, sunset angle and latitude of the power station under test for each day within the prediction period. Divide the amount of radiation received at the atmospheric surface of the power station under test by the total horizontal radiation of the power station on the same day to obtain the clear sky identification parameters for the corresponding day. Based on the sunny day identification parameters, sunny days within the prediction period are determined, and the entire 24 hours of non-sunny days are defined as the shadow period for all inverters of the power station. Determine the hourly average DC current dispersion rate of each inverter during sunny days within the forecast period; When the hourly average DC current dispersion rate exceeds a preset threshold, the corresponding hour is determined as the shadow period of the corresponding inverter. Determine all shadow periods for each inverter within the prediction period; The atmospheric surface radiation received by the power station is determined based on the daily declination angle, sunset angle, and latitude of the power station under test within the prediction period, including the following steps: The daily declination angle within the prediction period is determined by the following formula: ， ， in, Indicates the first i The solar declination angle of the day, with parameters a, b, c, d, e, f, and g, are derived from the Spencer operator. i This indicates the ordinal number of the natural day within the calendar year in which the prediction period is located. Indicates the sun angle; The sunset hour angle for each day within the prediction period is determined by the following formula: ,in Indicates the first i At sunset, Indicates the latitude of the power station to be measured. Represents pi (π). Represents the inverse cosine function; The following formula is used to determine the number of power stations to be measured within the prediction period. i The amount of radiation received by the surface of the sun's atmosphere. ,in, Indicates the power station under test. i The amount of radiation received by the surface of the sun's atmosphere. A Represents the solar constant. Indicates the power station under test. i The correction value caused by the distance between the Sun and the Earth.

7. A computer storage medium, characterized in that, The computer storage medium stores instructions, which, when executed, implement the photovoltaic power loss prediction method based on big data for photovoltaic power plants according to any one of claims 1 to 4, or the photovoltaic power plant shadow identification method based on big data according to claim 6.

Images