Distributed photovoltaic power prediction method and device

Abstract

The invention relates to the technical field of new energy power generation, and particularly provides a distributed photovoltaic power prediction method and device, and the method comprises the steps: obtaining the prediction day weather forecast data of N numerical weather forecast grid points with the minimum distance from a distributed photovoltaic power station, and N is a preset value; determining prediction day weather forecast data of the distributed photovoltaic power station based on the prediction day weather forecast data of the N numerical weather forecast grid points and a pre-trained optimization weighting model; and taking the forecast day weather forecast data of the distributed photovoltaic power station as the input of a pre-trained weather-power mapping relation model to obtain the forecast day power data of the distributed photovoltaic power station output by the pre-trained weather-power mapping relation model. According to the technical scheme provided by the invention, the characteristic of distributed photovoltaic scattered distribution can be considered, information of numerical weather forecast is fully utilized, universal application of a prediction algorithm to distributed photovoltaic in a region is realized, and thus the distributed photovoltaic power prediction precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy power generation, and particularly to a distributed photovoltaic power prediction method and device. Background Art

[0002] In recent years, photovoltaic power prediction technology has been gradually developed and matured, and photovoltaic power prediction systems have also been practically applied, playing an important role in formulating daily power generation plans, promoting new energy consumption, and ensuring the economic and safe operation of power grids.

[0003] At the present stage, the centralized photovoltaic power prediction technology has been relatively mature, mainly based on fixed-point, fixed-time, and quantitative numerical weather forecasts as inputs to construct a meteorological-power mapping relationship model. Different from centralized photovoltaics, distributed photovoltaics have the characteristics of many points, wide areas, and scattered distributions. At the present stage, distributed photovoltaic power prediction mainly adopts the method of centralized photovoltaic conversion or conversion near the reference station, without considering the spatial continuity of meteorological resources and the spatial distance between distributed photovoltaic power stations and reference stations, which restricts the improvement of the accuracy of distributed photovoltaic power prediction.

[0004] Therefore, how to consider the characteristics of the scattered distribution of distributed photovoltaics, make full use of the information of numerical weather forecasts, realize the universal applicability of the prediction algorithm to distributed photovoltaics in the region, and then improve the accuracy of distributed photovoltaic power prediction is an important problem faced by distributed photovoltaic power prediction. Summary of the Invention

[0005] In order to overcome the above defects, the present invention proposes a distributed photovoltaic power prediction method and device.

[0006] In a first aspect, a distributed photovoltaic power prediction method is provided, and the distributed photovoltaic power prediction method includes:

[0007] Obtain the predicted-day meteorological forecast data of the N numerical weather forecast grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value;

[0008] Determine the predicted-day meteorological forecast data of the distributed photovoltaic power station based on the predicted-day meteorological forecast data of the N numerical weather forecast grid points and a pre-trained optimization weighting model;

[0009] Use the predicted-day meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model to obtain the predicted-day power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

[0010] Preferably, the distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather forecast grid point, and the meteorological forecast data includes at least one of the following: total radiation, temperature.

[0011] Preferably, determining the predicted-day meteorological forecast data of the distributed photovoltaic power station based on the predicted-day meteorological forecast data of the N numerical weather prediction grid points and the pre-trained optimization weighting model includes:

[0012] Normalize the predicted-day meteorological forecast data of the numerical weather prediction grid points to obtain the normalized data corresponding to the predicted-day meteorological forecast data of the numerical weather prediction grid points;

[0013] Substitute the normalized data corresponding to the predicted-day meteorological forecast data of the numerical weather prediction grid points into the pre-trained optimization weighting model and solve to obtain the normalized data corresponding to the predicted-day meteorological forecast data of the distributed photovoltaic power station;

[0014] Denormalize the normalized data corresponding to the predicted-day meteorological forecast data of the distributed photovoltaic power station to obtain the predicted-day meteorological forecast data of the distributed photovoltaic power station.

[0015] Further, the normalized data corresponding to the predicted-day meteorological forecast data of the numerical weather prediction grid points is as follows:

[0016]

[0017] In the above formula, Y is the normalized data corresponding to the predicted-day meteorological forecast data of the numerical weather prediction grid points, X is the predicted-day meteorological forecast data of the numerical weather prediction grid points, and X min is the minimum value of the predicted-day meteorological forecast data of the numerical weather prediction grid points, and X max is the maximum value of the predicted-day meteorological forecast data of the numerical weather prediction grid points;

[0018] The predicted-day meteorological forecast data of the distributed photovoltaic power station is as follows:

[0019] X1 = Y1 × (X 1max - X 1min ) + X 1min

[0020] In the above formula, X1 is the predicted-day meteorological forecast data of the distributed photovoltaic power station, Y1 is the normalized data corresponding to the predicted-day meteorological forecast data of the distributed photovoltaic power station, X 1max is the maximum value of the predicted-day meteorological forecast data of the distributed photovoltaic power station, and X 1min is the minimum value of the predicted-day meteorological forecast data of the distributed photovoltaic power station.

[0021] Further, the structure of the pre-trained optimization weighting model is as follows:

[0022] P = β1P1+...+β N P N

[0023] In the above formula, P is the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, and β N is the weight of the Nth numerical weather prediction grid point, and P N is the normalized data corresponding to the meteorological forecast data of the Nth numerical weather prediction grid point.

[0024] Furthermore, the process of obtaining the weights of each numerical weather prediction grid point includes:

[0025] Constructing a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, and taking the minimum prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station as the goal, and iteratively calculating by the least square method.

[0026] Preferably, the pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. In the construction process, the training sample includes: the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained by using the pre-trained optimization weighting model.

[0027] In a second aspect, a distributed photovoltaic power prediction device is provided. The distributed photovoltaic power prediction device includes:

[0028] An acquisition module, configured to acquire the predicted-day meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value;

[0029] A first analysis module, configured to determine the predicted-day meteorological forecast data of the distributed photovoltaic power station based on the predicted-day meteorological forecast data of the N numerical weather prediction grid points and the pre-trained optimization weighting model;

[0030] A second analysis module, configured to use the predicted-day meteorological forecast data of the distributed photovoltaic power station as the input of the pre-trained meteorological-power mapping relationship model to obtain the predicted-day power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

[0031] Preferably, the distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather prediction grid point, and the meteorological forecast data includes at least one of the following: total radiation, temperature.

[0032] Preferably, the first analysis module is specifically configured to:

[0033] Normalize the meteorological forecast data of the predicted day for the numerical weather prediction grid points to obtain the normalized data corresponding to the meteorological forecast data of the predicted day for the numerical weather prediction grid points;

[0034] Substitute the normalized data corresponding to the meteorological forecast data of the predicted day for the numerical weather prediction grid points into the pre-trained optimization weighting model and solve to obtain the normalized data corresponding to the meteorological forecast data of the predicted day for the distributed photovoltaic power station;

[0035] Perform inverse normalization on the normalized data corresponding to the meteorological forecast data of the predicted day for the distributed photovoltaic power station to obtain the meteorological forecast data of the predicted day for the distributed photovoltaic power station.

[0036] Further, the normalized data corresponding to the meteorological forecast data of the predicted day for the numerical weather prediction grid points is as follows:

[0037]

[0038] In the above formula, Y is the normalized data corresponding to the meteorological forecast data of the predicted day for the numerical weather prediction grid points, X is the meteorological forecast data of the predicted day for the numerical weather prediction grid points, and X min is the minimum value of the meteorological forecast data of the predicted day for the numerical weather prediction grid points, and X max is the maximum value of the meteorological forecast data of the predicted day for the numerical weather prediction grid points;

[0039] The meteorological forecast data of the predicted day for the distributed photovoltaic power station is as follows:

[0040] X1 = Y1 × (X 1max - X 1min ) + X 1min

[0041] In the above formula, X1 is the meteorological forecast data of the predicted day for the distributed photovoltaic power station, Y1 is the normalized data corresponding to the meteorological forecast data of the predicted day for the distributed photovoltaic power station, and X 1max is the maximum value of the meteorological forecast data of the predicted day for the distributed photovoltaic power station, and X 1min is the minimum value of the meteorological forecast data of the predicted day for the distributed photovoltaic power station.

[0042] Further, the structure of the pre-trained optimization weighting model is as follows:

[0043] P = β1P1 +... + β N P N

[0044] In the above formula, P is the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, and βN is the weight of the Nth numerical weather prediction grid point, P N is the normalized data corresponding to the meteorological forecast data of the Nth numerical weather prediction grid point.

[0045] Furthermore, the process of obtaining the weights of each numerical weather prediction grid point includes:

[0046] Construct a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station. With the goal of minimizing the prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, iterative calculation is performed using the least squares method.

[0047] Preferably, the pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. The training sample in the construction process includes: the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained using the pre-trained optimization weighting model.

[0048] In a third aspect, a computer device is provided, including: one or more processors;

[0049] The processor is used to store one or more programs;

[0050] When the one or more programs are executed by the one or more processors, the distributed photovoltaic power prediction method described above is implemented.

[0051] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the distributed photovoltaic power prediction method described above is implemented.

[0052] One or more of the above technical solutions of the present invention have at least one or more of the following beneficial effects:

[0053] The present invention provides a distributed photovoltaic power prediction method and device, including: obtaining the predicted-day meteorological forecast data of N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value; determining the predicted-day meteorological forecast data of the distributed photovoltaic power station based on the predicted-day meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model; using the predicted-day meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model to obtain the predicted-day power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model. The technical solution provided by the present invention realizes the calculation of the forecast meteorological resources of the target site through the optimization weighting of adjacent grid points of grid-based numerical weather forecasting, and further realizes the power prediction modeling of the target distributed photovoltaic site. This technical solution can realize the power prediction calculation of any distributed photovoltaic site within the coverage of numerical weather forecasting, with high calculation efficiency, strong adaptability, and stable calculation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a schematic flowchart of the main steps of the distributed photovoltaic power prediction method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The following further elaborates in detail the specific embodiments of the present invention with reference to the accompanying drawings.

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0057] As disclosed in the background art, in recent years, photovoltaic power prediction technology has gradually developed and matured, and photovoltaic power prediction systems have also been practically applied, playing an important role in formulating daily power generation plans, promoting the consumption of new energy, and ensuring the economic and safe operation of the power grid.

[0058] At the present stage, the centralized photovoltaic power prediction technology has been relatively mature, mainly based on fixed-point, fixed-time, and quantitative numerical weather forecasts as inputs to construct a meteorological-power mapping relationship model. Different from centralized photovoltaics, distributed photovoltaics have the characteristics of numerous points, wide distribution, and scattered distribution. At the present stage, distributed photovoltaic power prediction mainly adopts the methods of centralized photovoltaic conversion or conversion based on adjacent reference stations, without considering the spatial continuity of meteorological resources and the spatial distance between distributed photovoltaic power stations and reference stations, which restricts the improvement of distributed photovoltaic power prediction accuracy.

[0059] Therefore, how to consider the characteristics of the decentralized distribution of distributed photovoltaic power generation, make full use of the information of numerical weather prediction, achieve the universal applicability of the prediction algorithm to distributed photovoltaic power generation within the region, and then improve the prediction accuracy of distributed photovoltaic power, is an important issue faced by distributed photovoltaic power prediction.

[0060] To solve the above problems, the present invention provides a method and device for predicting distributed photovoltaic power, including: obtaining the predicted daily meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value; determining the predicted daily meteorological forecast data of the distributed photovoltaic power station based on the predicted daily meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model; using the predicted daily meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model to obtain the predicted daily power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model. The technical solution provided by the present invention realizes the calculation of the predicted meteorological resources of the target site through the optimization weighting of adjacent grid points of grid-based numerical weather prediction, and then realizes the power prediction modeling of the target distributed photovoltaic site. This technical solution can realize the power prediction calculation of any distributed photovoltaic site within the coverage of numerical weather prediction, with high calculation efficiency, strong adaptability, and stable calculation results.

[0061] The above solution will be elaborated in detail below.

[0062] Embodiment 1

[0063] Refer to the appendix Figure 1 , Figure 1 which is a schematic diagram of the main step flow of the distributed photovoltaic power prediction method according to an embodiment of the present invention. As Figure 1 shown, the distributed photovoltaic power prediction method in the embodiment of the present invention mainly includes the following steps:

[0064] Step S101: Obtain the predicted daily meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value;

[0065] Step S102: Determine the predicted daily meteorological forecast data of the distributed photovoltaic power station based on the predicted daily meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model;

[0066] Step S103: Use the predicted daily meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model to obtain the predicted daily power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

[0067] Wherein, the distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather prediction grid point, and the weather forecast data includes at least one of the following: global radiation, temperature.

[0068] In this embodiment, determining the predicted-day weather forecast data of the distributed photovoltaic power station based on the predicted-day weather forecast data of the N numerical weather prediction grid points and the pre-trained optimization weighting model includes:

[0069] Performing normalization processing on the predicted-day weather forecast data of the numerical weather prediction grid point to obtain the normalized data corresponding to the predicted-day weather forecast data of the numerical weather prediction grid point;

[0070] Substituting the normalized data corresponding to the predicted-day weather forecast data of the numerical weather prediction grid point into the pre-trained optimization weighting model and solving to obtain the normalized data corresponding to the predicted-day weather forecast data of the distributed photovoltaic power station;

[0071] Performing inverse normalization processing on the normalized data corresponding to the predicted-day weather forecast data of the distributed photovoltaic power station to obtain the predicted-day weather forecast data of the distributed photovoltaic power station.

[0072] In one embodiment, the normalized data corresponding to the predicted-day weather forecast data of the numerical weather prediction grid point is as follows:

[0073]

[0074] In the above formula, Y is the normalized data corresponding to the predicted-day weather forecast data of the numerical weather prediction grid point, X is the predicted-day weather forecast data of the numerical weather prediction grid point, and X min is the minimum value of the predicted-day weather forecast data of the numerical weather prediction grid point, and X max is the maximum value of the predicted-day weather forecast data of the numerical weather prediction grid point;

[0075] The predicted-day weather forecast data of the distributed photovoltaic power station is as follows:

[0076] X1 = Y1 × (X 1max - X 1min ) + X 1min

[0077] In the above formula, X1 is the predicted-day weather forecast data of the distributed photovoltaic power station, Y1 is the normalized data corresponding to the predicted-day weather forecast data of the distributed photovoltaic power station, and X 1max is the maximum value of the predicted-day weather forecast data of the distributed photovoltaic power station, and X 1min is the minimum value of the predicted-day weather forecast data of the distributed photovoltaic power station.

[0078] In one embodiment, the structure of the pre-trained optimization weighting model is as follows:

[0079] P = β1P1+...+β N P N

[0080] In the above formula, P is the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, and β N is the weight of the Nth numerical weather prediction grid point, and P N is the normalized data corresponding to the meteorological forecast data of the Nth numerical weather prediction grid point.

[0081] In one embodiment, the process of obtaining the weights of each numerical weather prediction grid point includes:

[0082] Construct a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station. With the goal of minimizing the prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, iterative calculation is performed using the least squares method.

[0083] In this embodiment, the pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. The training samples in the construction process include: the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained using the pre-trained optimization weighting model.

[0084] For example: taking the data of the above training sample set as input, calculate the historical total radiation and temperature forecast data of the target site, and construct a one-to-many meteorological-power mapping relationship model using a support vector machine; taking the multi-grid point total radiation and temperature forecast data of the day to be predicted as input, calculate the power prediction result of the target site on the day to be predicted.

[0085] Embodiment 2

[0086] Based on the same inventive concept, the present invention also provides a distributed photovoltaic power prediction device, which includes:

[0087] An acquisition module, configured to acquire the predicted-day meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value;

[0088] A first analysis module, configured to determine the predicted-day meteorological forecast data of the distributed photovoltaic power station based on the predicted-day meteorological forecast data of the N numerical weather prediction grid points and the pre-trained optimization weighting model;

[0089] A second analysis module, configured to use the predicted daily meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model, and obtain the predicted daily power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

[0090] Preferably, the distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather prediction grid point, and the meteorological forecast data includes at least one of the following: total radiation, temperature.

[0091] Preferably, the first analysis module is specifically configured to:

[0092] Normalize the predicted daily meteorological forecast data of the numerical weather prediction grid point to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point;

[0093] Substitute the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point into a pre-trained optimization weighting model and solve to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station;

[0094] Denormalize the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station to obtain the predicted daily meteorological forecast data of the distributed photovoltaic power station.

[0095] Furthermore, the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point is as follows:

[0096]

[0097] In the above formula, Y is the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point, X is the predicted daily meteorological forecast data of the numerical weather prediction grid point, X min is the minimum value of the predicted daily meteorological forecast data of the numerical weather prediction grid point, X max is the maximum value of the predicted daily meteorological forecast data of the numerical weather prediction grid point;

[0098] The predicted daily meteorological forecast data of the distributed photovoltaic power station is as follows:

[0099] X1 = Y1×(X 1max - X 1min ) + X 1min

[0100] In the above formula, X1 is the predicted daily meteorological forecast data of the distributed photovoltaic power station, Y1 is the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station, X1max is the maximum value of the predicted daily meteorological forecast data for the distributed photovoltaic power station, X 1min is the minimum value of the predicted daily meteorological forecast data for the distributed photovoltaic power station.

[0101] Furthermore, the structure of the pre-trained optimization weighting model is as follows:

[0102] P = β1P1 +... + β N P N

[0103] In the above formula, P is the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, β N is the weight of the Nth numerical weather prediction grid point, and P N is the normalized data corresponding to the meteorological forecast data of the Nth numerical weather prediction grid point.

[0104] Furthermore, the process of obtaining the weights of each numerical weather prediction grid point includes:

[0105] Construct a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, and use the least squares method to iteratively calculate with the goal of minimizing the prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station.

[0106] Preferably, the pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. The training sample in the construction process includes: the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained using the pre-trained optimization weighting model.

[0107] Example 3

[0108] Based on the same inventive concept, the present invention further provides a computer device, which includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a distributed photovoltaic power prediction method in the above embodiments.

[0109] Embodiment 4

[0110] Based on the same inventive concept, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and, of course, the extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. And, one or more instructions suitable for being loaded and executed by the processor are also stored in this storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The one or more instructions stored in the computer-readable storage medium can be loaded and executed by the processor to implement the steps of a distributed photovoltaic power prediction method in the above embodiments.

[0111] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0112] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0113] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0114] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable devices provide steps for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent substitutions, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A distributed photovoltaic power prediction method, characterized in that, The method includes: Obtaining the predicted daily meteorological forecast data of N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value; Determining the predicted daily meteorological forecast data of the distributed photovoltaic power station based on the predicted daily meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model; Using the predicted daily meteorological forecast data of the distributed photovoltaic power station as the input of a pre-trained meteorological-power mapping relationship model to obtain the predicted daily power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

2. The method according to claim 1, wherein The distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather prediction grid point, and the meteorological forecast data includes at least one of the following: total radiation, temperature.

3. The method according to claim 1, characterized in that, The determining the predicted daily meteorological forecast data of the distributed photovoltaic power station based on the predicted daily meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model includes: Performing normalization processing on the predicted daily meteorological forecast data of the numerical weather prediction grid points to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid points; Substituting the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid points into the pre-trained optimization weighting model and solving to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station; Performing inverse normalization processing on the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station to obtain the predicted daily meteorological forecast data of the distributed photovoltaic power station.

4. The method according to claim 3, wherein The normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid points is as follows: In the above formula, Y is the normalized data corresponding to the meteorological forecast data of the prediction day at the numerical weather prediction grid point, X is the meteorological forecast data of the prediction day at the numerical weather prediction grid point, and X min is the minimum value of the meteorological forecast data of the prediction day at the numerical weather prediction grid point, and X max is the maximum value of the meteorological forecast data of the prediction day at the numerical weather prediction grid point; The predicted daily meteorological forecast data of the distributed photovoltaic power station is as follows: X1 = Y1×(X 1max -X 1min ) + X 1min In the above formula, X1 is the predicted daily meteorological forecast data of the distributed photovoltaic power station, Y1 is the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station, X 1max is the maximum value of the predicted daily meteorological forecast data of the distributed photovoltaic power station, X 1min is the minimum value of the predicted daily meteorological forecast data of the distributed photovoltaic power station.

5. The method according to claim 3, wherein The structure of the pre-trained optimization weighting model is as follows: P = β1P1+...+β N P N In the above formula, P is the normalized data corresponding to the weather forecast data of the distributed photovoltaic power station, and β N is the weight of the Nth numerical weather prediction grid point, and P N is the normalized data corresponding to the weather forecast data of the Nth numerical weather prediction grid point.

6. The method according to claim 5, characterized in that The process of obtaining the weights of each numerical weather prediction grid point includes: Constructing a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, and iteratively calculating using the least squares method with the goal of minimizing the prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station.

7. The method according to claim 1, wherein The pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. The training sample in the construction process includes: the historical meteorological forecast data of N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained using the pre-trained optimization weighting model.

8. A distributed photovoltaic power prediction device, characterized in that, The device includes: An acquisition module, configured to acquire the predicted daily meteorological forecast data of N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, where N is a preset value; A first analysis module, configured to determine the predicted daily meteorological forecast data of the distributed photovoltaic power station based on the predicted daily meteorological forecast data of the N numerical weather prediction grid points and a pre-trained optimization weighting model; A second analysis module, configured to use the predicted daily meteorological forecast data of the distributed photovoltaic power station as an input to a pre-trained meteorological-power mapping relationship model, and obtain the predicted daily power data of the distributed photovoltaic power station output by the pre-trained meteorological-power mapping relationship model.

9. The device according to claim 8, characterized in that, The distance is the Euclidean distance between the distributed photovoltaic power station and the numerical weather prediction grid point, and the meteorological forecast data includes at least one of the following: global radiation, temperature.

10. The device according to claim 8, characterized in that, The first analysis module is specifically configured to: Perform normalization processing on the predicted daily meteorological forecast data of the numerical weather prediction grid point to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point; Substitute the normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point into a pre-trained optimization weighting model and solve it to obtain the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station; Perform inverse normalization processing on the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station to obtain the predicted daily meteorological forecast data of the distributed photovoltaic power station.

11. The device according to claim 10, wherein The normalized data corresponding to the predicted daily meteorological forecast data of the numerical weather prediction grid point is as follows: In the above formula, Y is the normalized data corresponding to the meteorological forecast data of the predicted day at the numerical weather prediction grid point, X is the meteorological forecast data of the predicted day at the numerical weather prediction grid point, and X min is the minimum value of the meteorological forecast data of the predicted day at the numerical weather prediction grid point, and X max is the maximum value of the meteorological forecast data of the predicted day at the numerical weather prediction grid point; The predicted daily meteorological forecast data of the distributed photovoltaic power station is as follows: X1 = Y1×(X 1max - X 1min ) + X 1min In the above formula, X1 is the predicted daily meteorological forecast data of the distributed photovoltaic power station, Y1 is the normalized data corresponding to the predicted daily meteorological forecast data of the distributed photovoltaic power station, X 1max is the maximum value of the predicted daily meteorological forecast data of the distributed photovoltaic power station, X 1min is the minimum value of the predicted daily meteorological forecast data of the distributed photovoltaic power station.

12. The device according to claim 10, characterized in that, The structure of the pre-trained optimization weighting model is as follows: P = β1P1+...+β N P N In the above formula, P is the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station, and β N is the weight of the Nth numerical weather prediction grid point, and P N is the normalized data corresponding to the meteorological forecast data of the Nth numerical weather prediction grid point.

13. The device according to claim 12, characterized in that, The process of obtaining the weights of each numerical weather prediction grid point includes: Constructing a training sample with the normalized data corresponding to the historical meteorological forecast data of the distributed photovoltaic power station and the normalized data corresponding to the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station, and using the least squares method to iteratively calculate with the goal of minimizing the prediction deviation of the normalized data corresponding to the meteorological forecast data of the distributed photovoltaic power station.

14. The device according to claim 8, characterized in that, The pre-trained meteorological-power mapping relationship model is constructed using the support vector machine algorithm. In the construction process, the training sample includes: the historical meteorological forecast data of the N numerical weather prediction grid points with the smallest distance from the distributed photovoltaic power station and the historical meteorological forecast data of the distributed photovoltaic power station obtained using the pre-trained optimization weighting model.

15. A computer device, characterized in that, Including: One or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the distributed photovoltaic power prediction method according to any one of claims 1 to 7 is implemented.

16. A computer-readable storage medium, characterized in that, There is a computer program stored thereon, and when the computer program is executed, the distributed photovoltaic power prediction method according to any one of claims 1 to 7 is implemented.

Images