Power generation data prediction method and device based on meteorological data
By using alignment and correlation modeling methods, the problem of inaccurate correlation between meteorological data and power generation data was solved, enabling accurate prediction and high-quality monitoring of photovoltaic module power generation data.
Patent Information
- Application Number
- CN202511297117.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-11
- Publication Date
- 2026-01-16
AI Technical Summary
Existing technologies make it difficult to establish an effective correlation between meteorological data and power generation data, resulting in inaccurate predictions of photovoltaic module power generation data and affecting high-quality monitoring.
By acquiring historical power generation data and historical meteorological data of photovoltaic modules, aligning the data with a unified timestamp, establishing a correlation model, and using forecast meteorological data to predict power generation data, including time correction and data filtering to improve accuracy.
It enables accurate prediction of photovoltaic module power generation data, ensuring high-quality monitoring results.
Smart Images
Figure CN121355869A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of power generation forecasting technology, and in particular to a method and apparatus for forecasting power generation data based on meteorological data. Background Technology
[0002] A photovoltaic (PV) module is a power generation device that uses semiconductor materials (such as silicon) to convert light energy into direct current (DC). Its core principle is based on the photovoltaic effect, achieving stable power generation through a solid-state structure with no moving parts. Different meteorological data have different impacts on PV modules, necessitating the establishment of an effective correlation between meteorological data and power generation data to ensure accurate prediction of PV module power generation data and thus achieve high-quality monitoring of PV modules. Summary of the Invention
[0003] This disclosure aims to at least partially address one of the technical problems in the related art.
[0004] Therefore, the purpose of this disclosure is to provide a method and apparatus for predicting power generation data based on meteorological data.
[0005] To achieve the above objectives, the first aspect of this disclosure provides a method for predicting power generation data based on meteorological data, comprising: acquiring historical power generation data of photovoltaic modules and historical meteorological data of the area where the photovoltaic modules are located; using a unified timestamp to align the time of the historical power generation data and the historical meteorological data; filtering the historical power generation data and the historical meteorological data, and establishing a database of the historical power generation data and the historical meteorological data; establishing a correlation model between the power generation data and the meteorological data based on the database; acquiring future predicted meteorological data of the area where the photovoltaic modules are located, and obtaining future predicted power generation data of the photovoltaic modules based on the predicted meteorological data and the correlation model.
[0006] Optionally, the method further includes: obtaining power generation abrupt change points in the historical power generation data and meteorological abrupt change points in the historical meteorological data; obtaining the time difference between the historical power generation data and the historical meteorological data based on the time corresponding to the power generation abrupt change points and the time corresponding to the meteorological abrupt change points; and correcting the time of the historical power generation data and / or the time of the historical meteorological data according to the time difference to align the time of the historical power generation data and the historical meteorological data.
[0007] Optionally, the method further includes: acquiring multiple power generation abrupt change points in the historical power generation data and multiple meteorological abrupt change points in the historical meteorological data; determining multiple sets of corresponding power generation abrupt change points and meteorological abrupt change points; obtaining the average time difference between the historical power generation data and the historical meteorological data based on the time corresponding to the power generation abrupt change point and the time corresponding to the meteorological abrupt change point in each set; and correcting the time of the historical power generation data and / or the time of the historical meteorological data based on the average time difference to align the time of the historical power generation data and the historical meteorological data.
[0008] Optionally, the method further includes: sorting multiple power generation mutation points according to the time of the historical power generation data, and sorting multiple meteorological mutation points according to the time of the historical meteorological data; determining multiple sets of corresponding power generation mutation points and meteorological mutation points based on the sorting positions of the power generation mutation points and the meteorological mutation points.
[0009] Optionally, the method further includes: deploying a power generation data monitoring device at the power generation end of the photovoltaic module and deploying a meteorological data monitoring device in the area where the photovoltaic module is located; using a unified time, acquiring the historical meteorological data of the area where the photovoltaic module is located using the meteorological data monitoring device, and acquiring the historical power generation data of the photovoltaic module using the power generation data monitoring device.
[0010] Optionally, the method further includes: setting a power generation data threshold for the historical power generation data and a meteorological data threshold for the historical meteorological data; removing abnormal power generation data from the historical power generation data based on the power generation data threshold; and removing abnormal meteorological data from the historical meteorological data based on the meteorological data threshold.
[0011] Optionally, the method further includes: establishing a minute-level correlation model between power generation data and meteorological data based on the database and setting the time interval precision to minutes; obtaining minute-level predicted meteorological data for the area where the photovoltaic module is located; and obtaining minute-level predicted power generation data for the photovoltaic module based on the predicted minute-level meteorological data and the minute-level correlation model; establishing an hour-level correlation model between power generation data and meteorological data based on the database and setting the time interval precision to hours; obtaining hour-level predicted meteorological data for the area where the photovoltaic module is located; and obtaining hour-level predicted power generation data for the photovoltaic module based on the predicted hour-level meteorological data and the hour-level correlation model; and establishing a monthly correlation model between power generation data and meteorological data based on the database and setting the time interval precision to months; obtaining monthly predicted meteorological data for the area where the photovoltaic module is located; and obtaining monthly predicted power generation data for the photovoltaic module based on the predicted monthly meteorological data and the monthly correlation model.
[0012] Optionally, the method further includes: reacquiring historical power generation data of the photovoltaic module and historical meteorological data of the area where the photovoltaic module is located at preset intervals; updating the correlation model based on the reacquiring historical power generation data and the historical meteorological data; and obtaining future predicted power generation data of the photovoltaic module based on the predicted meteorological data and the updated correlation model.
[0013] Optionally, the method further includes: developing a visual forecast image based on the predicted power generation data and the meteorological data.
[0014] A second aspect of this disclosure provides a power generation data prediction device based on meteorological data, comprising: a data acquisition module for acquiring historical power generation data of a photovoltaic module and historical meteorological data of the area where the photovoltaic module is located; a preprocessing module for aligning the historical power generation data and the historical meteorological data using a unified timestamp, filtering the historical power generation data and the historical meteorological data, and establishing a database of the historical power generation data and the historical meteorological data; and a processing module for establishing a correlation model between the power generation data and the meteorological data based on the database, acquiring future predicted meteorological data of the area where the photovoltaic module is located, and obtaining future predicted power generation data of the photovoltaic module based on the predicted meteorological data and the correlation model.
[0015] The technical solution provided in this disclosure may include the following beneficial effects: By using a database to establish a correlation model between power generation data and meteorological data, and by obtaining the predicted power generation data of photovoltaic modules based on the predicted meteorological data and the correlation model, accurate prediction of photovoltaic module power generation data is achieved, thereby ensuring high-quality monitoring of photovoltaic modules.
[0016] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description
[0017] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which: Figure 1 This is a flowchart illustrating a power generation data prediction method based on meteorological data proposed in one embodiment of this disclosure. Detailed Implementation
[0018] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are used only to explain this disclosure, and should not be construed as limiting this disclosure. Rather, embodiments of this disclosure include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.
[0019] like Figure 1 As shown in the embodiments of this disclosure, a method for predicting power generation data based on meteorological data is proposed, including: S1: Obtain historical power generation data of photovoltaic modules and historical meteorological data of the area where the photovoltaic modules are located; S2: Use a unified timestamp to align the time of historical power generation data and historical meteorological data; S3: Filter historical power generation data and historical meteorological data, and establish a database of historical power generation data and historical meteorological data; S4: Establish a correlation model between power generation data and meteorological data based on the database; S5: Obtain future forecast meteorological data for the area where the photovoltaic modules are located, and obtain future forecast power generation data for the photovoltaic modules based on the forecast meteorological data and the correlation model.
[0020] It is understandable that acquiring historical power generation data of photovoltaic (PV) modules and historical meteorological data of the region where the PV modules are located, and aligning the historical power generation data with the historical meteorological data over time, can ensure an effective correspondence between the historical power generation data and the historical meteorological data. Furthermore, by filtering the historical power generation data and historical meteorological data, the accuracy of the database based on these data can be improved. Based on this, a correlation model between power generation data and meteorological data can be established using the database, and the predicted power generation data of the PV modules can be obtained based on the predicted meteorological data and the correlation model. This enables accurate prediction of the power generation data of the PV modules, thereby ensuring high-quality monitoring of the PV modules.
[0021] It should be noted that the meteorological data for photovoltaic modules can be set according to actual needs, and there are no restrictions on this. For example, meteorological data includes: solar radiation data, temperature data, atmospheric condition data, weather phenomenon data, wind condition data, etc. Specifically, solar radiation data includes: total horizontal radiation (total solar radiation received by a horizontal plane), direct normal radiation (radiation perpendicular to the direction of the sun beam), diffused radiation (radiation that reaches the ground after being scattered by the atmosphere), module tilt radiation (radiation actually received by the module after considering the installation tilt angle), ultraviolet radiation index (ultraviolet band intensity that affects the aging of encapsulation materials), etc.; temperature data includes: ambient air temperature (air temperature around the module), module backsheet temperature (measured or estimated cell operating temperature), and ground surface temperature (temperature that affects the aging of encapsulation materials). Local microclimate data includes: ground temperature; atmospheric conditions data include: relative humidity (humidity affecting the rate of surface dirt accumulation), atmospheric pressure (altitude correction parameter affecting air density and heat dissipation), aerosol optical thickness (particulate matter concentration index characterizing atmospheric turbidity), and precipitable water (atmospheric water vapor content affecting solar radiation attenuation); weather phenomenon data includes: cloud cover (total cloud cover / low cloud cover ratio), precipitation type and intensity (real-time data of rain / snow / hail), snow cover thickness (snow cover depth on module surface), and dew condensation (duration of dew on module surface); wind conditions data include: wind speed (horizontal wind speed at module height), wind direction (angle of oncoming wind relative to module installation direction), and turbulence intensity (wind field fluctuation index affecting module mechanical stress).
[0022] The power generation data of photovoltaic modules can be set according to actual needs without limitation. For example, power generation data includes: power output data, electrical characteristic data, operating status data, system response data, etc. Specifically, power output data includes: DC power (real-time DC output of the string / array), AC power (active power output from the inverter to the grid), power generation (cumulative energy generated within a time period), conversion efficiency (ratio of actual output to theoretical maximum output), etc.; electrical characteristic data includes: IV curve parameters (open-circuit voltage, short-circuit current, maximum power point), fill factor, series / parallel resistance, etc.; operating status data includes: operating temperature (cell temperature estimated by sensors or algorithms), shading ratio (shading coverage area ratio identified by intelligent algorithms), aging degradation rate (annual degradation rate compared to initial output), fault codes (abnormal status codes reported by the inverter / monitoring system), etc.; system response data includes: MPPT tracking efficiency (actual efficiency of the maximum power point tracking algorithm), start-up / shutdown sequence (daily system wake-up and sleep time records), power limiting situation (proportion of active load reduction due to grid dispatch), etc.
[0023] Regarding the correlation between meteorological data and power generation data, for example, when the backsheet temperature of a photovoltaic module rises by 10°C, the open-circuit voltage of the photovoltaic module drops by approximately 21mV, and the conversion efficiency decreases by 4% at a rate of 1000W / m². 2 Radiation causes a reduction of DC power of about 3%-5%; when the snow cover reaches 2cm, the radiation received by the slope drops to 10%-20% of the normal value, and the DC power output is usually 5% lower than the rated value, accompanied by an abnormal increase in operating temperature (snow insulation effect).
[0024] The timelines used for collecting historical power generation data and historical meteorological data are prone to discrepancies, which can lead to a lack of effective correspondence between the two. However, by using time alignment, the accurate establishment of the correlation model is ensured.
[0025] Historical power generation and meteorological data contain a large number of abnormal data (values that are too large or too small) and missing data (data from some time points is lost or not collected). Therefore, by filtering historical power generation and meteorological data, abnormal data can be removed, ensuring the accurate establishment of the database and improving the efficiency of subsequent data processing.
[0026] The correlation model between power generation data and meteorological data establishes an effective correspondence between them. Different meteorological data can lead to different power generation data, and the correlation model can achieve multi-data fusion and correlation between various data.
[0027] In some embodiments, the method further includes: Obtain power generation abrupt changes in historical power generation data and meteorological abrupt changes in historical meteorological data; Based on the time corresponding to the power generation mutation point and the time corresponding to the meteorological mutation point, the time difference between historical power generation data and historical meteorological data is obtained. The time of historical power generation data and / or the time of historical meteorological data are corrected according to the time difference to align the time of historical power generation data and historical meteorological data.
[0028] It is understandable that, based on the correspondence between power generation mutation points in historical power generation data and meteorological mutation points in historical meteorological data, the time difference between historical power generation data and historical meteorological data can be obtained. This time difference can then be used to correct the time of historical power generation data and / or the time of historical meteorological data, thereby aligning the time of historical power generation data and historical meteorological data and ensuring an effective correspondence between them.
[0029] It should be noted that power generation abrupt changes in historical power generation data can be sudden increases or decreases in certain power generation parameters, just as meteorological abrupt changes in historical meteorological data can be sudden increases or decreases in certain meteorological parameters. For example, a sudden drop in irradiance intensity can cause a sudden drop in power generation. Based on the correlation between power generation data and meteorological data, and by utilizing the correspondence between abrupt events in historical power generation data and historical meteorological data, time alignment of historical power generation data and historical meteorological data can be effectively achieved.
[0030] In some embodiments, the method further includes: Obtain multiple power generation abrupt change points from historical power generation data and multiple meteorological abrupt change points from historical meteorological data; Multiple sets of corresponding power generation abrupt change points and meteorological abrupt change points were identified; Based on the time corresponding to the power generation mutation point and the time corresponding to the meteorological mutation point in each group, the average time difference between historical power generation data and historical meteorological data is obtained. The timing of historical power generation data and / or historical meteorological data is corrected based on the average time difference to align the timing of historical power generation data and historical meteorological data.
[0031] Understandably, based on multiple sets of corresponding power generation abrupt change points and meteorological abrupt change points, multiple time differences can be obtained. Furthermore, based on these multiple time differences, the average time difference between historical power generation data and historical meteorological data can be obtained, thereby achieving precise time alignment between historical power generation data and historical meteorological data using the average time difference.
[0032] It should be noted that the time difference between power generation abrupt change points and meteorological abrupt change points is used for time correction and alignment between historical power generation data and historical meteorological data. Multiple sets of power generation abrupt change points and meteorological abrupt change points provide more time differences for correction. Using the average of multiple time differences for time correction further improves the accuracy of time alignment.
[0033] In some embodiments, the method further includes: Multiple power generation abrupt changes were sorted based on the time of historical power generation data, and multiple meteorological abrupt changes were sorted based on the time of historical meteorological data. Multiple sets of corresponding power generation abrupt change points and meteorological abrupt change points were determined based on the sorting positions of power generation abrupt change points and meteorological abrupt change points.
[0034] Understandably, by sorting multiple power generation abrupt changes and multiple meteorological abrupt changes according to the timeline, the corresponding power generation abrupt changes and meteorological abrupt changes can be determined. Then, based on multiple sets of corresponding power generation abrupt changes and meteorological abrupt changes, the average time difference between historical power generation data and historical meteorological data can be obtained, ensuring accurate time alignment.
[0035] It should be noted that when consecutive abrupt events occur, the time interval between adjacent abrupt events is short, which can easily lead to incorrect correspondence between power generation abrupt events and meteorological abrupt events. However, the method of sorting the correspondence along the timeline can accurately identify the corresponding power generation abrupt events and meteorological abrupt events.
[0036] In some embodiments, the method further includes: Power generation data monitoring devices are installed at the power generation end of the photovoltaic modules, and meteorological data monitoring devices are installed in the area where the photovoltaic modules are located. A unified timeframe is used to acquire historical meteorological data for the area where the photovoltaic modules are located using meteorological data monitoring devices, and historical power generation data for the photovoltaic modules is acquired using power generation data monitoring devices.
[0037] It is understandable that by using a unified time period and acquiring historical meteorological data of the area where the photovoltaic modules are located using meteorological data monitoring devices, and acquiring historical power generation data of the photovoltaic modules using power generation data monitoring devices, a rough correspondence between historical meteorological data and historical power generation data can be achieved, thereby significantly reducing the difficulty of time alignment between historical meteorological data and historical power generation data.
[0038] It should be noted that the power generation data monitoring device is used to monitor the power generation data of photovoltaic modules. The specific type of power generation data monitoring device can be set according to actual needs, and there are no restrictions on it.
[0039] Meteorological data monitoring devices are used to monitor meteorological data in the area where photovoltaic modules are located. The specific type of meteorological data monitoring device can be set according to actual needs and there are no restrictions on it.
[0040] Even when using a unified time, there will still be a small time discrepancy between historical meteorological data obtained by meteorological data monitoring devices and historical power generation data obtained by power generation data monitoring devices, so time alignment is necessary.
[0041] In some embodiments, the method further includes: Set thresholds for historical power generation data and historical meteorological data; Abnormal power generation data in historical power generation data is removed based on power generation data thresholds, and abnormal meteorological data in historical meteorological data is removed based on meteorological data thresholds.
[0042] It is understandable that removing abnormal power generation data from historical power generation data based on power generation data thresholds, and removing abnormal meteorological data from historical meteorological data based on meteorological data thresholds, can effectively improve the accuracy of databases based on historical power generation data and historical meteorological data, thereby ensuring accurate prediction of photovoltaic module power generation data.
[0043] It should be noted that different power generation parameters correspond to different power generation data thresholds, and different meteorological data correspond to different meteorological data thresholds. The specific threshold can be set according to actual needs, and there are no restrictions on it.
[0044] In some embodiments, the method further includes: Based on the database and with the time interval precision set to minutes, a minute-level correlation model between power generation data and meteorological data is established. In addition, minute-level predicted meteorological data of the area where the photovoltaic module is located is obtained. Based on the predicted minute-level meteorological data and the minute-level correlation model, minute-level predicted power generation data of the photovoltaic module is obtained. Based on the database and with the time interval precision set to hours, an hourly correlation model between power generation data and meteorological data is established. Hourly predicted meteorological data for the area where the photovoltaic modules are located is obtained. Hourly predicted power generation data for the photovoltaic modules is obtained based on the predicted hourly meteorological data and the hourly correlation model. Based on the database and with the time interval precision set to monthly, a monthly correlation model between power generation data and meteorological data is established. In addition, monthly predicted meteorological data of the area where the photovoltaic modules are located is obtained. Based on the predicted monthly meteorological data and the monthly correlation model, monthly predicted power generation data of the photovoltaic modules is obtained.
[0045] Understandably, minute-level predicted power generation data for photovoltaic (PV) modules is obtained based on predicted minute-level meteorological data and minute-level correlation models, thus achieving minute-level power generation data prediction; hour-level predicted power generation data is obtained based on predicted hourly meteorological data and hourly correlation models, thus achieving hourly power generation data prediction; and monthly predicted power generation data is obtained based on predicted monthly meteorological data and monthly correlation models, thus achieving monthly power generation data prediction. This ensures high-quality monitoring of PV modules.
[0046] It should be noted that minute-level power generation data forecasts can characterize the short-term fluctuations of photovoltaic modules, hour-level power generation data forecasts can characterize the daily changes of photovoltaic modules, and monthly-level power generation data forecasts can characterize the seasonal changes of photovoltaic modules.
[0047] In some embodiments, the method further includes: Historical power generation data of photovoltaic modules and historical meteorological data of the area where photovoltaic modules are located are reacquired at preset intervals. The correlation model is updated based on the reacquired historical power generation data and historical meteorological data; Predicted future power generation data for photovoltaic modules is obtained based on forecasted meteorological data and updated correlation models.
[0048] It is understandable that the historical power generation data of the photovoltaic modules and the historical meteorological data of the area where the photovoltaic modules are located are reacquired at preset intervals, and the correlation model is updated based on the reacquired historical power generation data and historical meteorological data, so as to ensure the accurate prediction of photovoltaic module power generation data.
[0049] It should be noted that updates based on a preset frequency can continuously supplement new historical data, thereby constantly correcting the correlation model and improving prediction accuracy.
[0050] In some embodiments, the method further includes: Visualized forecast images are created based on predicted power generation data and meteorological data.
[0051] Understandably, visual forecast images are created based on predicted power generation data and meteorological data, which can then be used to display the predicted power generation data more intuitively, making it easier for operators to monitor photovoltaic modules.
[0052] This embodiment also proposes a power generation data prediction device based on meteorological data, including: an acquisition module, a preprocessing module, and a processing module. The acquisition module is used to acquire historical power generation data of photovoltaic modules and historical meteorological data of the area where the photovoltaic modules are located. The preprocessing module is used to align the time of historical power generation data and historical meteorological data using a unified timestamp, and to filter historical power generation data and historical meteorological data and establish a database of historical power generation data and historical meteorological data. The processing module is used to establish a correlation model between power generation data and meteorological data based on the database, and to acquire future predicted meteorological data of the area where the photovoltaic modules are located, and to obtain future predicted power generation data of the photovoltaic modules based on the predicted meteorological data and the correlation model.
[0053] It is understandable that acquiring historical power generation data of photovoltaic (PV) modules and historical meteorological data of the region where the PV modules are located, and aligning the historical power generation data with the historical meteorological data over time, can ensure an effective correspondence between the historical power generation data and the historical meteorological data. Furthermore, by filtering the historical power generation data and historical meteorological data, the accuracy of the database based on these data can be improved. Based on this, a correlation model between power generation data and meteorological data can be established using the database, and the predicted power generation data of the PV modules can be obtained based on the predicted meteorological data and the correlation model. This enables accurate prediction of the power generation data of the PV modules, thereby ensuring high-quality monitoring of the PV modules.
[0054] It should be noted that in the description of this disclosure, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this disclosure, unless otherwise stated, "a plurality of" means two or more.
[0055] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.
[0056] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0057] Although embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present disclosure.
Claims
1. A power generation data prediction method based on meteorological data, characterized by, The method comprises: obtaining historical power generation data of a photovoltaic module and historical meteorological data of an area where the photovoltaic module is located; aligning the time of the historical power generation data and the historical meteorological data by using a unified timestamp; screening the historical power generation data and the historical meteorological data and establishing a database of the historical power generation data and the historical meteorological data; establishing a correlation model of power generation data and meteorological data based on the database; obtaining future predicted meteorological data of the area where the photovoltaic module is located, and obtaining future predicted power generation data of the photovoltaic module according to the predicted meteorological data and the correlation model. 2.The method of claim 1, wherein, The method further comprises: obtaining power generation mutation points in the historical power generation data and meteorological mutation points in the historical meteorological data; obtaining the time difference between the historical power generation data and the historical meteorological data based on the time corresponding to the power generation mutation points and the time corresponding to the meteorological mutation points; correcting the time of the historical power generation data and / or the time of the historical meteorological data according to the time difference, so as to align the time of the historical power generation data and the historical meteorological data. 3.The method of claim 2, wherein, The method further comprises: obtaining a plurality of power generation mutation points in the historical power generation data and a plurality of meteorological mutation points in the historical meteorological data; determining a plurality of groups of corresponding power generation mutation points and meteorological mutation points; obtaining the average time difference between the historical power generation data and the historical meteorological data according to the time corresponding to the power generation mutation points and the time corresponding to the meteorological mutation points in each group; correcting the time of the historical power generation data and / or the time of the historical meteorological data according to the average time difference, so as to align the time of the historical power generation data and the historical meteorological data. 4.The method of claim 3, wherein, The method further comprises: sorting a plurality of the power generation mutation points according to the time of the historical power generation data, and sorting a plurality of the meteorological mutation points according to the time of the historical meteorological data; determining a plurality of groups of corresponding power generation mutation points and meteorological mutation points based on the sorted positions of the power generation mutation points and the sorted positions of the meteorological mutation points. 5.The method of claim 1, wherein, The method further comprises: arranging a power generation data monitoring device at a power generation end of the photovoltaic module, and arranging a meteorological data monitoring device in the area where the photovoltaic module is located; obtaining the historical meteorological data of the area where the photovoltaic module is located by using the meteorological data monitoring device and obtaining the historical power generation data of the photovoltaic module by using the power generation data monitoring device by using a unified time. 6.The method of claim 1, wherein, The method further comprises: setting a power generation data threshold of the historical power generation data and a meteorological data threshold of the historical meteorological data; removing abnormal power generation data in the historical power generation data based on the power generation data threshold, and removing abnormal meteorological data in the historical meteorological data based on the meteorological data threshold. 7.The method of claim 1, wherein, The method further comprises: establishing a minute-level correlation model of power generation data and meteorological data based on the database and setting a time interval precision as minute, and obtaining minute-level predicted meteorological data of a region where the photovoltaic module is located, and obtaining minute-level predicted power generation data of the photovoltaic module according to the predicted minute-level meteorological data and the minute-level correlation model; establishing an hour-level correlation model of power generation data and meteorological data based on the database and setting a time interval precision as hour, and obtaining hour-level predicted meteorological data of the region where the photovoltaic module is located, and obtaining hour-level predicted power generation data of the photovoltaic module according to the predicted hour-level meteorological data and the hour-level correlation model; establishing a month-level correlation model of power generation data and meteorological data based on the database and setting a time interval precision as month, and obtaining month-level predicted meteorological data of the region where the photovoltaic module is located, and obtaining month-level predicted power generation data of the photovoltaic module according to the predicted month-level meteorological data and the month-level correlation model. 8.The method of claim 1, wherein, The method further comprises: renewing the historical power generation data of the photovoltaic module and the historical meteorological data of the region where the photovoltaic module is located every preset time; updating the correlation model based on the renewed historical power generation data and the historical meteorological data; obtaining future predicted power generation data of the photovoltaic module according to the predicted meteorological data and the updated correlation model. 9.The method of claim 1, wherein, The method further comprises: making a visual prediction image according to the predicted power generation data and the meteorological data.
10. A power generation data prediction device based on meteorological data, characterized by, comprises: a collection module, which is used to obtain historical power generation data of a photovoltaic module and historical meteorological data of a region where the photovoltaic module is located; a preprocessing module, which is used to align times of the historical power generation data and the historical meteorological data by using a uniform timestamp, and to filter the historical power generation data and the historical meteorological data and establish a database of the historical power generation data and the historical meteorological data; a processing module, which is used to establish a correlation model of power generation data and meteorological data based on the database, and to obtain future predicted meteorological data of the region where the photovoltaic module is located, and to obtain future predicted power generation data of the photovoltaic module according to the predicted meteorological data and the correlation model.