A distributed photovoltaic cluster output prediction method and device considering temporal and spatial characteristics

Abstract

The present invention relates to a distributed photovoltaic cluster output prediction method and device considering time and space characteristics, wherein the method comprises: collecting historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; extracting the illumination time, illumination angle and solar irradiance on inclined surfaces from the historical weather data using a photovoltaic physics model; using the illumination time, illumination angle, solar irradiance on inclined surfaces, ambient temperature, ground wind speed and historical photovoltaic production data as new data sets; training a BP-ANN model using the new data set to obtain a distributed photovoltaic cluster output prediction model; inputting the illumination time, illumination angle, solar irradiance on inclined surfaces, ambient temperature and ground wind speed of the day to be predicted into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted. The present invention can improve the accuracy of photovoltaic power generation prediction at a lower cost.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation cluster prediction, and in particular to a distributed photovoltaic cluster output prediction method and device taking into account time and space characteristics. Background Art

[0002] With the construction of new power systems, a large number of distributed photovoltaics are connected to the power system. As we all know, the output state of photovoltaics cannot be fully adjusted or arranged in advance, but depends on weather conditions such as sunshine. Since changes in power generation and power quality will have an adverse impact on the stability and reliability of the entire system and the service quality of users, power generation needs to be predicted during the operation of the power system. For this reason, solar photovoltaic power generation cluster prediction is crucial to improving the overall performance and reliability of the power system and improving the operation planning of grid operators. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a distributed photovoltaic cluster output prediction method and device taking into account the temporal and spatial characteristics, which can improve the accuracy of photovoltaic power generation prediction at a lower cost.

[0004] The technical solution adopted by the present invention to solve the technical problem is: to provide a distributed photovoltaic cluster output prediction method considering the temporal and spatial characteristics, comprising the following steps:

[0005] Collecting historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; the historical weather data includes hourly counter data, daily counter data, global radiation on the horizontal plane, ambient temperature and ground wind speed;

[0006] A photovoltaic physics model is used to extract the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data;

[0007] The illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed and historical photovoltaic production data are used as a new data set;

[0008] The BP-ANN model is trained using the new data set to obtain a distributed photovoltaic cluster output prediction model;

[0009] The illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted are input into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

[0010] Before extracting the illumination time, illumination angle and solar irradiance on the inclined surface from the historical weather data using the photovoltaic physics model, the method further includes:

[0011] The collected historical photovoltaic output data and historical weather data are preprocessed, and the preprocessing includes zero-value data removal processing, noise data removal processing and normalization processing.

[0012] The method of extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data using the photovoltaic physics model specifically includes:

[0013] Obtaining sunlight time based on the location of the distributed photovoltaic cluster, hour counter data, and day counter data;

[0014] Calculate the hour angle based on the hour counter data;

[0015] Calculate the inclination angle based on the daily counter data;

[0016] Calculate the elevation angle of the sun in the sky observed from the horizontal plane according to the hour angle and the position of the distributed photovoltaic cluster;

[0017] Calculate the solar azimuth angle according to the inclination angle, the hour angle and the elevation angle, wherein the solar azimuth angle is the illumination angle;

[0018] Calculate the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster based on the inclination angle and the hour angle;

[0019] The solar irradiance on the inclined surface is calculated based on the angle between the straight lines on the surface of the distributed photovoltaic cluster.

[0020] The hour angle is calculated by HRA=15°×(LST-12), wherein HRA is the hour angle and LST is the solar time at the location of the distributed photovoltaic cluster.

[0021] The inclination angle is Calculated, where δ is the inclination angle and d is the daily counter data.

[0022] The elevation angle is Calculated, where α is the elevation angle; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle, and HRA is the hour angle.

[0023] The solar azimuth is given by Calculated, where γ is the solar azimuth; δ is the inclination; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. HRA is the hour angle. α is the elevation angle.

[0024] The angle between the sunlight and the straight line perpendicular to the surface of the distributed photovoltaic cluster is It is calculated as follows, where θ is the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value, and when the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle. β is the surface inclination angle of the distributed photovoltaic cluster. Zs is the surface azimuth. When the distributed photovoltaic cluster is located in the western hemisphere, the surface azimuth is a positive value, and when the distributed photovoltaic cluster is located in the eastern hemisphere, the surface azimuth is a negative value. HRA is the hour angle.

[0025] The solar irradiance on the inclined surface is given by Calculated, where G t is the solar irradiance on the inclined surface, R B is the ratio of the beam radiation on the inclined surface to the beam radiation on the horizontal surface, G B is the beam radiation on the horizontal surface, G D is the diffuse radiation on the horizontal surface, β is the tilt angle of the distributed photovoltaic cluster surface, ρ G is the ground albedo.

[0026] The technical solution adopted by the present invention to solve the technical problem is: to provide a distributed photovoltaic cluster output prediction device considering the temporal and spatial characteristics, comprising:

[0027] A collection module, used to collect historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; the historical weather data includes hourly counter data, daily counter data, global radiation on the horizontal plane, ambient temperature and ground wind speed;

[0028] An extraction module, for extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data by using a photovoltaic physics model;

[0029] A new data set module, for taking the illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed and historical photovoltaic output data as a new data set;

[0030] The training module is used to train the BP-ANN model using the new data set to obtain a distributed photovoltaic cluster output prediction model;

[0031] The prediction module is used to input the sunlight time, sunlight angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

[0032] The technical solution adopted by the present invention to solve its technical problem is: to provide an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned distributed photovoltaic cluster output prediction method considering the temporal and spatial characteristics when executing the computer program.

[0033] The technical solution adopted by the present invention to solve its technical problem is: providing a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned distributed photovoltaic cluster output prediction method considering the temporal and spatial characteristics are implemented.

[0034] Beneficial Effects

[0035] Due to the adoption of the above-mentioned technical scheme, the present invention has the following advantages and positive effects compared with the prior art: the present invention converts and extracts characteristic information related to the spatiotemporal characteristics from the original weather data, and reconstructs a new data set with these characteristic information and historical photovoltaic production data, and uses the new data set to design an accurate prediction model that takes into account the spatiotemporal characteristics, thereby improving the accuracy of photovoltaic power generation prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a distributed photovoltaic cluster output prediction method considering time and space characteristics according to a first embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the sun's declination at different times;

[0038] Figure 3 This is a schematic diagram of various angles of the sun;

[0039] Figure 4 It is a comparison chart of the prediction results of the method of this embodiment and the conventional method. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the appended claims of the application equally.

[0041] The first embodiment of the present invention relates to a distributed photovoltaic cluster output prediction method that takes into account the spatiotemporal characteristics. The method converts and extracts feature information related to the spatiotemporal characteristics from the original weather data, and reconstructs a new data set with these feature information and historical photovoltaic output data, and uses the new data set to train the BP-ANN network model to obtain a distributed photovoltaic cluster output prediction model, and uses the distributed photovoltaic cluster output prediction model for prediction. The new data set can enhance the ability of the prediction model, thereby providing a more accurate photovoltaic power generation prediction at a lower computing cost. Figure 1 As shown, the distributed photovoltaic cluster output prediction method considering the spatiotemporal characteristics of this embodiment specifically includes the following steps:

[0042] Step 1: Collect historical photovoltaic output data and historical weather data of distributed photovoltaic clusters.

[0043] The data collected in this step mainly includes hour counter data and daily counter data Site horizontal radiation data Ambient temperature data Surface wind speed data and photovoltaic production data The above data are all in hourly units. The collected data represent the hourly weather variables that affect the PV system, as well as the corresponding timestamps. In this step, the PV output for three consecutive years is collected.

[0044] After the data is collected, it can be preprocessed. Since photovoltaic production data is also recorded in the early morning and at night, and the photovoltaic production data in these time periods is zero, the zero values ​​in the photovoltaic production data can be removed during preprocessing. Some negative values ​​are also recorded in the weather data, and these negative values ​​indicate that there is some noise in the output signal of the sensor, or the sensor may need calibration, so these noise data can be removed during preprocessing. In order to make the established BP-ANN prediction model run effectively, during preprocessing, the photovoltaic production data and weather data can also be normalized so that the data range of these data is between 0-1.

[0045] Step 2: extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data using a photovoltaic physics model.

[0046] This step collects the hour counter data from the raw data. Daily counter data Site horizontal radiation data Convert to light time Lighting Angle and solar irradiance on inclined surfaces The specific method is as follows:

[0047] Step A, obtaining the sunshine time according to the location of the distributed photovoltaic cluster, the hour counter data and the day counter data Figure 2 A schematic diagram of the sun's declination at different times is given. For the northern hemisphere, the sunshine time is the longest at the summer solstice and the shortest at the winter solstice.

[0048] Step B, calculate the hour angle based on the hour counter data. The hour angle HRA is the conversion of the local solar time (LST) into an angle that describes the position of the sun using the Earth's rotation rate (15° / hour). It should be noted that this angle is positive during the day and negative at night. It is calculated as follows:

[0049] HRA = 15° × (LST-12);

[0050] Among them, the solar time LST can be obtained based on the hour counter data.

[0051] Step C, calculate the inclination angle δ based on the day counter data. Due to the inclination angle of the earth to its own axis and the rotation of the earth around the sun, the inclination angle δ will change with the seasons (such as Figure 2 The inclination angle δ is calculated as:

[0052]

[0053] Here, d is the day counter data. Assuming that the current day is January 1, d=1.

[0054] Step D, calculate the elevation angle α of the sun in the sky observed from the horizontal plane according to the hour angle HRA and the position of the distributed photovoltaic cluster. The elevation angle α is calculated as follows:

[0055]

[0056] in, is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value.

[0057] Step E, calculate the zenith angle ζ according to the elevation angle α. The zenith angle ζ is defined as the angle between the sun and the vertical plane, which can be estimated by:

[0058] ζ=90°-α.

[0059] Step F, calculate the solar azimuth angle γ according to the inclination angle δ, the hour angle HRA and the elevation angle α, such as Figure 3 As shown, the solar azimuth angle γ is the illumination angle, which can be estimated by the following formula:

[0060]

[0061] Step G, calculating the angle θ between the sunlight and the straight line perpendicular to the surface of the distributed photovoltaic cluster according to the inclination angle δ and the hour angle HRA. The angle is the incident angle of the sunlight on the surface of the distributed photovoltaic cluster, which can be calculated by the following formula:

[0062]

[0063] Wherein, β is the surface inclination angle of the distributed photovoltaic cluster; Zs is the surface azimuth, when the distributed photovoltaic cluster is located in the western hemisphere, the surface azimuth is positive, when the distributed photovoltaic cluster is located in the eastern hemisphere, the surface azimuth is negative.

[0064] Step H, calculates the solar irradiance on the inclined surface based on the angle between the straight lines on the surface of the distributed photovoltaic cluster. The collectors of photovoltaic devices are usually installed at a certain angle rather than a plane to increase the amount of captured energy while reducing reflection and cosine losses (losses caused by the increase in the angle between solar radiation and the surface normal). Therefore, it is necessary to obtain solar radiation information on the inclined collector, but most of the radiation data recorded or calculated can only be obtained on horizontal surfaces. This step can convert the data obtained on these horizontal surfaces into radiation data on the inclined surface. The solar irradiance on the inclined surface can be calculated by the following formula:

[0065]

[0066] Among them, G t is the solar irradiance on the inclined surface, R B G is the ratio of the beam radiation on the inclined surface to the beam radiation on the horizontal surface. This value can be calculated by the inclination angle of the distributed photovoltaic cluster surface and the incident angle of sunlight on the distributed photovoltaic cluster surface. B is the beam radiation on the horizontal surface, G D is the diffuse radiation on the horizontal surface, β is the tilt angle of the distributed photovoltaic cluster surface, ρ G is the ground albedo, which is assumed to be 0.2 in this embodiment.

[0067] In step 3, the obtained illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed, and historical photovoltaic production data are used as a new data set.

[0068] Step 4, use the new data set to train the BP-ANN model to obtain a distributed photovoltaic cluster output prediction model, specifically: divide the new data set into a training data set and a test data set, use the training data set to train the BP-ANN model, and use the test data set to test the trained BP-ANN model. If the trained BP-ANN model meets the test requirements, the trained BP-ANN model is used as the distributed photovoltaic cluster output prediction model; if the trained BP-ANN model does not meet the test requirements, continue to use the training data set to train the BP-ANN model.

[0069] In this step, during training, the illumination time, illumination angle, solar irradiance on inclined surfaces, ambient temperature and ground wind speed in the training data set are used as inputs of the BP-ANN model. The output of the BP-ANN model is the predicted photovoltaic production data. The training goal is to minimize the difference between the predicted photovoltaic production data and the historical photovoltaic production data. Therefore, the difference between the predicted photovoltaic production data and the historical photovoltaic production data is used as the loss function to adjust the parameters of the BP-ANN model.

[0070] Step 5, input the illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

[0071] To verify the effectiveness of this implementation, the prediction results obtained by using the method of this implementation are compared with the prediction results obtained by using the original data and the prediction results obtained by using the features extracted by PCA. Figure 4 It can be seen that the method using this embodiment (blue line in the figure) performs better than using the original data (green line in the figure) and using the features extracted by PCA (red line in the figure).

[0072] It is not difficult to find that the present invention converts and extracts characteristic information related to the spatiotemporal characteristics from the original weather data, reconstructs a new data set with these characteristic information and historical photovoltaic production data, and uses the new data set to design an accurate prediction model that takes into account the spatiotemporal characteristics, thereby improving the accuracy of photovoltaic power generation prediction.

[0073] The second embodiment of the present invention relates to a distributed photovoltaic cluster output prediction device considering time and space characteristics, comprising:

[0074] A collection module, used to collect historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; the historical weather data includes hourly counter data, daily counter data, global radiation on the horizontal plane, ambient temperature and ground wind speed;

[0075] An extraction module, for extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data by using a photovoltaic physics model;

[0076] A new data set module, for taking the illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed and historical photovoltaic output data as a new data set;

[0077] A training module, used to divide the new data set into a training data set and a test data set, use the training data set to train the BP-ANN model, and use the test data set to test the trained BP-ANN model. If the trained BP-ANN model meets the test requirements, the trained BP-ANN model is used as a distributed photovoltaic cluster output prediction model; if the trained BP-ANN model does not meet the test requirements, the training data set is continued to be used to train the BP-ANN model;

[0078] The prediction module is used to input the sunlight time, sunlight angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

[0079] The distributed photovoltaic cluster output prediction device considering the spatiotemporal characteristics also includes: a preprocessing module, which is used to preprocess the collected historical photovoltaic output data and historical weather data, and the preprocessing includes removing zero-value data processing, removing noise data processing and normalization processing.

[0080] The extraction module comprises:

[0081] An acquisition unit, used for acquiring the sunshine time according to the location of the distributed photovoltaic cluster, the decimal counter data and the day counter data;

[0082] an hour angle calculation unit, for calculating the hour angle based on the hour counter data;

[0083] an inclination calculation unit, for calculating the inclination based on the day counter data;

[0084] An elevation angle calculation unit, used to calculate the elevation angle of the sun in the sky observed from the horizontal plane according to the hour angle and the position of the distributed photovoltaic cluster;

[0085] A solar azimuth angle calculation unit, used to calculate the solar azimuth angle according to the inclination angle, the hour angle and the elevation angle, wherein the solar azimuth angle is the illumination angle;

[0086] An incident angle calculation unit, used to calculate the angle between sunlight and a straight line perpendicular to the surface of a distributed photovoltaic cluster according to the inclination angle and the hour angle;

[0087] The solar irradiance calculation unit on the inclined surface is used to calculate the solar irradiance on the inclined surface according to the angle between the straight lines on the surface of the distributed photovoltaic cluster.

[0088] The hour angle calculation unit calculates the hour angle by HRA=15°×(LST-12), wherein HRA is the hour angle and LST is the solar time at the location of the distributed photovoltaic cluster.

[0089] The inclination angle calculation unit is Calculate the inclination angle, where δ is the inclination angle and d is the day counter data.

[0090] The elevation angle calculation unit is Calculate the elevation angle, where α is the elevation angle; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle, and HRA is the hour angle.

[0091] The solar azimuth angle calculation unit is Calculate the solar azimuth, where γ is the solar azimuth and δ is the inclination; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. HRA is the hour angle. α is the elevation angle.

[0092] The incident angle calculation unit is Calculate the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster, where θ is the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value, and when the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle. β is the surface inclination angle of the distributed photovoltaic cluster. Zs is the surface azimuth. When the distributed photovoltaic cluster is located in the western hemisphere, the surface azimuth is a positive value, and when the distributed photovoltaic cluster is located in the eastern hemisphere, the surface azimuth is a negative value. HRA is the hour angle.

[0093] The solar irradiance calculation unit on the inclined surface is Calculate the solar irradiance on a tilted surface, where G t is the solar irradiance on the inclined surface, R B is the ratio of the beam radiation on the inclined surface to the beam radiation on the horizontal surface, G B is the beam radiation on the horizontal surface, G Dis the diffuse radiation on the horizontal surface, β is the tilt angle of the distributed photovoltaic cluster surface, ρ G is the ground albedo.

[0094] A third embodiment of the present invention relates to an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for predicting the output of a distributed photovoltaic cluster taking into account the temporal and spatial characteristics of the first embodiment when executing the computer program.

[0095] The fourth embodiment of the present invention relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the distributed photovoltaic cluster output prediction method considering the spatiotemporal characteristics of the first embodiment.

[0096] 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 and optical storage, etc.) that contain computer-usable program code.

[0097] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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 generate 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.

[0098] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction method, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] 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.

[0100] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics, characterized in that: The following steps are involved: Collecting historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; the historical weather data includes hourly counter data, daily counter data, global radiation on the horizontal plane, ambient temperature and ground wind speed; A photovoltaic physics model is used to extract the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data; The illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed and historical photovoltaic production data are used as a new data set; The BP-ANN model is trained using the new data set to obtain a distributed photovoltaic cluster output prediction model; The illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted are input into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

2. The distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics according to claim 1 is characterized in that: Before extracting the illumination time, illumination angle and solar irradiance on the inclined surface from the historical weather data using the photovoltaic physics model, the method further includes: The collected historical photovoltaic output data and historical weather data are preprocessed, and the preprocessing includes zero-value data removal processing, noise data removal processing and normalization processing.

3. The distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics according to claim 1 is characterized in that: The method of extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data using the photovoltaic physics model specifically includes: Obtaining the sunshine time according to the location of the distributed photovoltaic cluster, the decimal counter data and the day counter data; Calculate the hour angle based on the hour counter data; Calculate the inclination angle based on the daily counter data; Calculate the elevation angle of the sun in the sky observed from the horizontal plane according to the hour angle and the position of the distributed photovoltaic cluster; Calculate the solar azimuth angle according to the inclination angle, the hour angle and the elevation angle, wherein the solar azimuth angle is the illumination angle; Calculate the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster based on the inclination angle and the hour angle; The solar irradiance on the inclined surface is calculated based on the angle between the straight lines on the surface of the distributed photovoltaic cluster.

4. The distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics according to claim 3 is characterized in that: The hour angle is calculated by HRA=15°×(LST-12), wherein HRA is the hour angle and LST is the solar time at the location of the distributed photovoltaic cluster.

5. The distributed photovoltaic cluster output prediction method considering time and space characteristics according to claim 3 is characterized in that: The inclination angle is Calculated, where δ is the inclination angle and d is the daily counter data.

6. The distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics according to claim 3 is characterized in that: The elevation angle is Calculated, where α is the elevation angle; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle, and HRA is the hour angle.

7. The distributed photovoltaic cluster output prediction method considering time and space characteristics according to claim 3 is characterized in that: The solar azimuth is given by Calculated, where γ is the solar azimuth; δ is the inclination; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. HRA is the hour angle. α is the elevation angle.

8. The distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics according to claim 3 is characterized in that: The angle between the sunlight and the straight line perpendicular to the surface of the distributed photovoltaic cluster is It is calculated as follows, where θ is the angle between sunlight and a straight line perpendicular to the surface of the distributed photovoltaic cluster; is the latitude of the distributed photovoltaic cluster. When the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a negative value, and when the distributed photovoltaic cluster is located in the southern hemisphere, the latitude is a positive value. δ is the inclination angle. β is the surface inclination angle of the distributed photovoltaic cluster. Zs is the surface azimuth. When the distributed photovoltaic cluster is located in the western hemisphere, the surface azimuth is a positive value, and when the distributed photovoltaic cluster is located in the eastern hemisphere, the surface azimuth is a negative value. HRA is the hour angle.

9. The distributed photovoltaic cluster output prediction method considering time and space characteristics according to claim 3 is characterized in that: The solar irradiance on the inclined surface is given by Calculated, where G t is the solar irradiance on the inclined surface, R B is the ratio of the beam radiation on the inclined surface to the beam radiation on the horizontal surface, G B is the beam radiation on the horizontal surface, G D is the diffuse radiation on the horizontal surface, β is the tilt angle of the distributed photovoltaic cluster surface, ρ G is the ground albedo.

10. A distributed photovoltaic cluster output prediction device considering temporal and spatial characteristics, characterized in that: include: A collection module, used to collect historical photovoltaic production data and historical weather data of distributed photovoltaic clusters; the historical weather data includes hourly counter data, daily counter data, global radiation on the horizontal plane, ambient temperature and ground wind speed; An extraction module, for extracting the sunshine time, sunshine angle and solar irradiance on the inclined surface from the historical weather data by using a photovoltaic physics model; A new data set module, for taking the illumination time, illumination angle, solar irradiance on the inclined surface, ambient temperature, ground wind speed and historical photovoltaic output data as a new data set; The training module is used to train the BP-ANN model using the new data set to obtain a distributed photovoltaic cluster output prediction model; The prediction module is used to input the sunlight time, sunlight angle, solar irradiance on the inclined surface, ambient temperature, and ground wind speed of the day to be predicted into the distributed photovoltaic cluster output prediction model to obtain the distributed photovoltaic cluster output data of the day to be predicted.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the distributed photovoltaic cluster output prediction method considering the temporal and spatial characteristics as described in any one of claims 1-9 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distributed photovoltaic cluster output prediction method considering spatiotemporal characteristics as described in any one of claims 1 to 9 are implemented.

Images