Refined weather prediction method and system based on photovoltaic range

Abstract

The invention discloses a fine meteorological prediction method and system based on a photovoltaic range, and belongs to the field of meteorological prediction.The method comprises the steps that a target photovoltaic network is determined, parameter ranges of a plurality of photovoltaic modules are collected respectively, and a photovoltaic range data set is constructed; acquiring historical power generation data and historical meteorological data according to the target photovoltaic network; establishing a virtual simulation platform based on the plurality of photovoltaic modules in combination with the photovoltaic range data set, performing association analysis of the power generation data and the meteorological data, and determining a power generation-meteorological mapping relationship; and according to the historical meteorological data, constructing a meteorological training sample set, constructing a meteorological prediction model, and providing meteorological prediction information for the target photovoltaic network. The technical problem that in the prior art, it is difficult to provide refined weather forecast information for a specific photovoltaic power station is solved, and the technical effect that finer weather forecast information is provided for the target photovoltaic power station, so that the photovoltaic power generation prediction accuracy is remarkably improved is achieved.

Description

Technical Field

[0001] The present invention relates to the field of weather forecasting, and in particular to a refined weather forecasting method and system based on photovoltaic range. Background Art

[0002] Photovoltaic power generation is developing rapidly as a clean and renewable energy form. However, photovoltaic power generation is greatly affected by meteorological conditions. Factors such as sunlight intensity, temperature, and wind speed directly determine the power generation efficiency of photovoltaic modules. Therefore, accurate prediction of the power generation of photovoltaic power stations is crucial for grid dispatching and power station operation management. Existing photovoltaic power generation prediction methods usually use methods based on historical data and empirical models. The meteorological forecast information is usually coarse-grained and cannot meet the needs of refined meteorological forecasts for photovoltaic power stations. Summary of the invention

[0003] The present application provides a photovoltaic range-based refined weather forecasting method and system, aiming to solve the technical problem in the prior art that it is difficult to provide refined weather forecast information for a specific photovoltaic power station.

[0004] In view of the above problems, the present application provides a refined meteorological forecasting method and system based on photovoltaic range.

[0005] The first aspect disclosed in the present application provides a refined meteorological forecasting method based on photovoltaic range, the method comprising: determining a target photovoltaic network, the target photovoltaic network comprising a plurality of photovoltaic modules, collecting parameter ranges of the plurality of photovoltaic modules respectively, and constructing a photovoltaic range data set; collecting historical power generation data according to the target photovoltaic network, and synchronously collecting corresponding historical meteorological data; building a virtual simulation platform based on the plurality of photovoltaic modules and the photovoltaic range data set; importing the historical power generation data and the historical meteorological data into the virtual simulation platform, performing correlation analysis between the power generation data and the meteorological data, and determining the power generation-meteorological mapping relationship; constructing a meteorological training sample set based on the power generation-meteorological mapping relationship and the historical meteorological data; and constructing a meteorological forecasting model using the meteorological training sample set to provide meteorological forecasting information for the target photovoltaic network.

[0006] Another aspect disclosed in the present application provides a refined meteorological forecasting system based on a photovoltaic range, the system comprising: a data set construction module, used to determine a target photovoltaic network, the target photovoltaic network comprising a plurality of photovoltaic components, and parameter ranges of the plurality of photovoltaic components are collected respectively to construct a photovoltaic range data set; a historical data collection module, used to collect historical power generation data according to the target photovoltaic network, and synchronously collect corresponding historical meteorological data; a simulation platform construction module, used to build a virtual simulation platform based on a plurality of photovoltaic components and in combination with the photovoltaic range data set; a mapping relationship determination module, used to import historical power generation data and historical meteorological data into the virtual simulation platform, perform correlation analysis between power generation data and meteorological data, and determine a power generation-meteorological mapping relationship; a sample construction module, used to construct a meteorological training sample set based on the power generation-meteorological mapping relationship and according to historical meteorological data; a meteorological prediction module, used to construct a meteorological prediction model using the meteorological training sample set, and provide meteorological prediction information for the target photovoltaic network.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The invention adopts the following technical solutions: determining the target photovoltaic network, collecting the parameter range of each photovoltaic component, and constructing a photovoltaic range data set; collecting historical power generation data and corresponding historical meteorological data according to the target photovoltaic network; building a virtual simulation platform based on the photovoltaic range data set, importing historical data into the platform, analyzing the relationship between power generation data and meteorological data, and determining the power generation-meteorological mapping relationship; constructing a meteorological training sample set based on the power generation-meteorological mapping relationship, and using the sample set to train a meteorological forecast model; applying the trained meteorological forecast model to the target photovoltaic network to provide it with more refined meteorological forecast information, which solves the technical problem in the prior art that it is difficult to provide refined meteorological forecast information for a specific photovoltaic power station, and achieves the technical effect of providing more refined meteorological forecast information for the target photovoltaic power station, thereby significantly improving the accuracy of photovoltaic power generation prediction.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic flow chart of a refined meteorological forecasting method based on photovoltaic range is provided for the embodiment of the present application;

[0011] Figure 2 A structural schematic diagram of a photovoltaic-based refined weather forecasting system is provided for an embodiment of the present application.

[0012] Explanation of the accompanying drawings: data set construction module 11, historical data collection module 12, simulation platform construction module 13, mapping relationship determination module 14, sample construction module 15, meteorological prediction module 16. DETAILED DESCRIPTION

[0013] The overall idea of ​​the technical solution provided by this application is as follows:

[0014] The embodiments of the present application provide a photovoltaic-based refined weather forecasting method and system.

[0015] First, determine the target photovoltaic network, collect the parameter range of each photovoltaic component, and construct a representative photovoltaic range data set. Secondly, collect historical power generation data and corresponding historical meteorological data according to the target photovoltaic network, and build a virtual simulation platform based on the photovoltaic range data set. By analyzing the relationship between power generation data and meteorological data, the mapping relationship between power generation and meteorology is established. Next, based on the power generation-meteorological mapping relationship, a meteorological training sample set is constructed, and the sample set is used to train a meteorological forecast model for the target photovoltaic network. Finally, the trained meteorological forecast model is applied to the target photovoltaic network to provide it with more refined meteorological forecast information, thereby effectively improving the accuracy of photovoltaic power generation prediction and meeting the needs of refined operation and management of photovoltaic power stations.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the embodiment of the present application provides a refined meteorological forecasting method based on photovoltaic range, the method comprising:

[0019] Determine a target photovoltaic network, the target photovoltaic network includes a plurality of photovoltaic components, collect parameter ranges of the plurality of photovoltaic components respectively, and construct a photovoltaic range data set;

[0020] Furthermore, this step includes:

[0021] Uniquely labeling the plurality of photovoltaic modules respectively, obtaining a plurality of photovoltaic module labels, and storing the plurality of photovoltaic module labels according to a list to construct a photovoltaic module list;

[0022] Traversing the photovoltaic component list, obtaining a first photovoltaic component, retrieving design parameters of the first photovoltaic component, and extracting a first parameter range from the design parameters according to a preset parameter dimension;

[0023] The first photovoltaic component is associated with the first parameter range and stored, and added to the photovoltaic range data set.

[0024] In the embodiment of the present application, first, determine the photovoltaic power station that needs to be refined for meteorological forecasting, and interact with the management system of the photovoltaic power station to obtain the target photovoltaic network, which includes multiple photovoltaic modules. Then, uniquely label the multiple photovoltaic modules in the target photovoltaic network respectively, so as to quickly and accurately identify and manage each photovoltaic module, and the label information of each photovoltaic module, and store these labels in order in the form of a list, and construct a photovoltaic module list to provide an effective photovoltaic module index and management means for subsequent steps. Subsequently, traverse the photovoltaic module list constructed above, obtain each photovoltaic module one by one, and retrieve the design parameter information of the photovoltaic module. Subsequently, for each photovoltaic module, extract the corresponding parameter range data from its design parameters according to the preset parameter dimensions (such as power generation, conversion efficiency, etc.), and fully obtain the specific value range of each photovoltaic module in each parameter dimension. Afterwards, the label of each photovoltaic module is associated with its corresponding parameter range data and stored, and added to the photovoltaic range data set. The photovoltaic range data set brings together the detailed parameter information of all photovoltaic modules in the target photovoltaic network, and provides the necessary input data support for the subsequent virtual simulation platform construction.

[0025] By determining the photovoltaic components of the target photovoltaic network and establishing a parameter range data set, a solid foundation is laid for subsequent virtual simulation and meteorological forecast model training.

[0026] According to the target photovoltaic network, historical power generation data is collected, and corresponding historical meteorological data is simultaneously collected;

[0027] In the embodiment of the present application, after the target photovoltaic network is determined, the historical power generation data of the target photovoltaic network is collected, and the corresponding historical meteorological data is collected synchronously.

[0028] First, for each photovoltaic module in the target photovoltaic network, its past actual power generation data records are collected. These historical power generation data records reflect the actual power generation status of each photovoltaic module in different time periods, providing basic data for the subsequent establishment of power generation-meteorological mapping relationship. At the same time, historical meteorological data corresponding to the above historical power generation data are collected. These historical meteorological data include the measurement values ​​of various meteorological parameters such as temperature, light, wind speed, etc., covering the main meteorological factors affecting photovoltaic power generation. By synchronously collecting historical power generation data and historical meteorological data, the relationship between the two is established, laying the necessary data foundation for subsequent data analysis and model training.

[0029] By obtaining the historical power generation data and corresponding historical meteorological data of the target photovoltaic network, support is provided for the subsequent construction of the virtual simulation platform, power generation-meteorological mapping relationship analysis, etc., laying the foundation for realizing refined meteorological forecasts.

[0030] Building a virtual simulation platform based on the plurality of photovoltaic components and in combination with the photovoltaic range data set;

[0031] Furthermore, this step includes:

[0032] Determine a photovoltaic module topology structure according to the target photovoltaic network, wherein the photovoltaic module topology structure includes a plurality of topology nodes corresponding to a plurality of photovoltaic modules one by one, and each topology node has a formatted photovoltaic module model;

[0033] Traversing a plurality of photovoltaic components in the photovoltaic range data set, determining a first photovoltaic component, extracting a first parameter range of the first photovoltaic component, and determining a first topological node corresponding to the first photovoltaic component in a photovoltaic component topological structure;

[0034] A formatted photovoltaic component model in the first topology node is set based on the first parameter range to obtain a virtual simulation platform.

[0035] In a feasible implementation, first, according to the determined target photovoltaic network, the topological structure of the photovoltaic network is determined. Specifically, the photovoltaic module topological structure includes a plurality of topological nodes, each of which corresponds to a specific photovoltaic module in the target photovoltaic network. At the same time, each topological node also has a formatted photovoltaic module model for simulating and characterizing the actual characteristics of the corresponding photovoltaic module. By constructing the photovoltaic module topological structure, the necessary structural foundation is laid for the virtual simulation platform in the subsequent steps. Then, all photovoltaic modules are traversed in the constructed photovoltaic range data set, and each photovoltaic module is determined one by one, and a photovoltaic module is obtained each time as the first photovoltaic module. For the first photovoltaic module, the first parameter range of the first photovoltaic module is extracted from the photovoltaic range data set. At the same time, in the photovoltaic module topological structure, the first topological node corresponding to the first photovoltaic module is determined, and the specific photovoltaic module parameter information is associated and corresponded with the node in the topological structure. Afterwards, based on the first parameter range of the first photovoltaic module, the formatted photovoltaic module model in the first topological node is set and configured so that the formatted photovoltaic module model can be simulated and modeled according to the actual parameter characteristics of the first photovoltaic module. By repeatedly configuring the remaining formatted photovoltaic component models, corresponding photovoltaic component models are configured for all nodes in the topology structure, thereby obtaining a complete virtual simulation platform.

[0036] By constructing a topological structure model of the target photovoltaic network and configuring the photovoltaic component model of each node according to the parameter information in the photovoltaic range data set, a platform for virtual simulation is obtained, providing necessary data support for the subsequent power generation-meteorological mapping relationship analysis.

[0037] Importing the historical power generation data and the historical meteorological data into a virtual simulation platform, performing correlation analysis between the power generation data and the meteorological data, and determining a power generation-meteorological mapping relationship;

[0038] Furthermore, this step includes:

[0039] Preprocessing the historical power generation data and the historical meteorological data to obtain a plurality of power generation data sequences and a plurality of meteorological data sequences, wherein the plurality of power generation data sequences correspond to the plurality of meteorological data sequences one by one;

[0040] According to multiple power generation data sequences and multiple meteorological data sequences, the meteorological parameters in the meteorological data are locked, the influence on the power generation data is analyzed, and the influence weight of the meteorological parameters is obtained;

[0041] A power generation-meteorological mapping relationship is obtained based on the meteorological parameter impact weights.

[0042] Furthermore, the meteorological parameter influence weights are obtained, including:

[0043] Traversing and extracting a first power generation data sequence and a first meteorological data sequence from a plurality of power generation data sequences and a plurality of meteorological data sequences;

[0044] In the virtual simulation platform, a first meteorological parameter is determined in the first meteorological data sequence, and other meteorological parameters are locked to obtain a first meteorological impact value of the first meteorological parameter on the first power generation data sequence;

[0045] A first meteorological parameter influence weight is obtained according to the first meteorological influence value, and is added to the meteorological parameter influence weight.

[0046] In a preferred embodiment, the historical power generation data and historical meteorological data collected above are preprocessed to extract multiple power generation data sequences and multiple meteorological data sequences from these original data, and ensure that these power generation data sequences and meteorological data sequences are one-to-one corresponding to ensure that a mapping relationship between power generation data and meteorological data can be established in subsequent analysis, thereby laying a good data foundation for subsequent power generation-meteorological relationship analysis.

[0047] Then, among multiple power generation data sequences and multiple meteorological data sequences, the first power generation data sequence and the first meteorological data sequence are traversed and extracted, and the two data sequences are one-to-one corresponding, that is, there is a direct corresponding relationship between the first power generation data sequence and the first meteorological data sequence. Subsequently, using the constructed virtual simulation platform, a specific first meteorological parameter is determined in the first meteorological data sequence, and the remaining meteorological parameters are locked, that is, the other meteorological parameters are kept unchanged. On this basis, the meteorological impact value of the first power generation data sequence corresponding to the first meteorological parameter is analyzed and calculated to obtain the first meteorological impact value, which reflects the degree of effect of a single meteorological parameter on photovoltaic power generation. Then, according to the first meteorological impact value of the first meteorological parameter obtained, the influence weight of the parameter is calculated to reflect the relative importance of the meteorological parameter in the entire meteorological system. Then, the first meteorological impact value is added to the meteorological parameter impact weight data, and the impact analysis results of the meteorological parameters are continuously supplemented and improved, which provides an important basis for establishing the power generation-meteorological mapping relationship and lays a key foundation for realizing refined meteorological forecasts.

[0048] Then, using the impact weights corresponding to each meteorological parameter, a quantitative power generation-meteorological correlation is established, where the coefficient corresponding to each meteorological parameter is its impact weight, reflecting the relative importance of the parameter to power generation. The power generation-meteorological mapping relationship is not static, but needs to be dynamically adjusted and optimized according to actual conditions. For example, when the photovoltaic system is upgraded or the operating environment changes, it is necessary to re-evaluate the impact weights of each meteorological parameter and update the power generation-meteorological correlation accordingly to ensure that the power generation-meteorological mapping relationship remains highly consistent and provide a reliable basis for meteorological forecasting.

[0049] By preprocessing and organizing the raw data, and then deeply analyzing the impact weights of meteorological parameters on power generation, a mapping relationship between power generation and meteorology is constructed, laying a key foundation for subsequent refined meteorological forecasts.

[0050] Based on the power generation-meteorological mapping relationship and according to the historical meteorological data, a meteorological training sample set is constructed;

[0051] Furthermore, this step includes:

[0052] Extracting the meteorological parameter impact weights in the power generation-meteorological mapping relationship, marking the historical meteorological data based on the meteorological parameter impact weights, and acquiring initial meteorological data;

[0053] Meteorological data without weight identification in the initial meteorological data are eliminated to obtain identified meteorological data, and a meteorological training sample set is obtained based on the identified meteorological data.

[0054] In a feasible implementation, first, the established power generation-meteorological mapping relationship is extracted, which contains the influence weight information of each meteorological parameter. Based on the influence weights of these meteorological parameters, the historical meteorological data is identified and screened. Specifically, the historical meteorological data is traversed, each meteorological parameter is identified according to its influence weight, the key meteorological parameters that have a relatively large impact on power generation are highlighted, and a set of identified initial meteorological data is obtained. Then, for the initial meteorological data, those meteorological parameter data without weight identification are further eliminated. Specifically, only those data identified as key meteorological parameters are retained to construct a set of identified meteorological data. Based on these identified meteorological data, a meteorological training sample set is obtained and assembled. This training sample set focuses on the key meteorological factors that have a greater impact on power generation, and provides targeted data support for subsequent meteorological forecast model training.

[0055] Starting from the power generation-meteorology mapping relationship established in the early stage, key meteorological parameter data are targeted and extracted to construct a highly targeted meteorological training sample set. This can not only improve the accuracy of the training model, but also greatly improve the reliability of subsequent meteorological forecasts, providing strong support for photovoltaic power generation forecast applications.

[0056] A meteorological prediction model is constructed using the meteorological training sample set to provide meteorological prediction information for the target photovoltaic network.

[0057] In the embodiment of the present application, after obtaining the meteorological training sample set, the constructed meteorological training sample set is used to further develop and train the meteorological prediction model. The meteorological prediction model can accurately predict future meteorological changes based on historical meteorological data and provide refined meteorological prediction information for the target photovoltaic network.

[0058] Specifically, a modeling method based on machine learning is used to train and optimize the prediction model using a meteorological training sample set. Through in-depth learning of a large amount of historical meteorological data, the prediction model can accurately capture the complex correlation between meteorological parameters and use these laws to predict future meteorological conditions. Then, the meteorological prediction model is used to perform meteorological prediction on the target photovoltaic network and obtain meteorological prediction information.

[0059] By developing a meteorological prediction model based on machine learning and using the established meteorological training samples for model training, we can provide refined meteorological forecast information for the target photovoltaic network, laying a solid foundation for the accurate prediction of photovoltaic power generation.

[0060] Furthermore, the method further comprises:

[0061] Obtaining the prediction accuracy of the weather forecast information;

[0062] When the prediction accuracy is lower than the accuracy threshold, the meteorological prediction model is optimized.

[0063] In a feasible implementation, the meteorological data actually observed by the target photovoltaic network is collected regularly, and compared and analyzed one by one with the meteorological forecast information given by the meteorological forecast model. By statistically analyzing the degree of deviation between the meteorological forecast information and the actual observed value, the prediction accuracy index of the meteorological forecast model is calculated to measure the meteorological forecast performance, which not only reflects the accuracy of the meteorological forecast model, but also characterizes its reliability and stability. Only when the prediction accuracy reaches a certain level can the meteorological forecast information truly provide practical support for the photovoltaic power generation business.

[0064] Then, the threshold requirement for prediction accuracy is specified as the accuracy threshold. Once the prediction accuracy of the meteorological prediction model is found to be lower than the accuracy threshold, optimization and improvement measures are immediately initiated to optimize the meteorological prediction model, including re-collecting and supplementing training sample data, adjusting the parameter settings of the prediction model, and introducing more advanced machine learning algorithms. Through continuous optimization, the accuracy of the meteorological prediction model is ensured, providing a reliable basic support for refined meteorological prediction.

[0065] By regularly evaluating and continuously optimizing meteorological forecast performance, we ensure that the prediction accuracy of the meteorological forecast model can be maintained at an ideal level for a long time, providing continuous, stable and reliable meteorological forecast support for photovoltaic power generation business.

[0066] In summary, the photovoltaic range-based refined weather forecasting method provided by the embodiment of the present application has the following technical effects:

[0067] 1. Provide more detailed weather forecast information for specific photovoltaic power stations, which is a significant improvement over traditional weather forecast information and provides strong support for the refined operation and management of photovoltaic power stations.

[0068] 2. Based on refined weather forecast information and combined with the power generation-weather mapping relationship, the power generation of photovoltaic power stations can be predicted more accurately, improving the accuracy of weather forecasts.

[0069] 3. By establishing a photovoltaic range virtual simulation platform, we can deeply analyze the intrinsic relationship between power generation characteristics and meteorological factors under different photovoltaic parameter conditions, providing important support for optimizing meteorological forecasts for photovoltaic power stations.

[0070] In summary, the refined meteorological forecasting method based on the photovoltaic range proposed in this application can effectively improve the accuracy of photovoltaic power generation prediction, support the refined operation of power stations, and provide support for the intelligent and refined operation and maintenance of photovoltaic power generation.

[0071] Embodiment 2

[0072] Based on the same inventive concept as the photovoltaic range-based refined meteorological forecasting method in the aforementioned embodiment, Figure 2 As shown, the embodiment of the present application provides a refined weather forecasting system based on photovoltaic range, the system comprising:

[0073] A data set construction module 11 is used to determine a target photovoltaic network, where the target photovoltaic network includes a plurality of photovoltaic components, and to collect parameter ranges of the plurality of photovoltaic components to construct a photovoltaic range data set;

[0074] A historical data collection module 12 is used to collect historical power generation data according to the target photovoltaic network, and simultaneously collect corresponding historical meteorological data;

[0075] A simulation platform building module 13, used to build a virtual simulation platform based on the plurality of photovoltaic components and in combination with the photovoltaic range data set;

[0076] A mapping relationship determination module 14 is used to import the historical power generation data and the historical meteorological data into the virtual simulation platform, perform correlation analysis between the power generation data and the meteorological data, and determine the power generation-meteorological mapping relationship;

[0077] A sample construction module 15, configured to construct a meteorological training sample set based on the power generation-meteorological mapping relationship and the historical meteorological data;

[0078] The meteorological prediction module 16 is used to construct a meteorological prediction model using the meteorological training sample set to provide meteorological prediction information for the target photovoltaic network.

[0079] Furthermore, the data set construction module 11 includes the following execution steps:

[0080] Uniquely labeling the plurality of photovoltaic modules respectively, obtaining a plurality of photovoltaic module labels, and storing the plurality of photovoltaic module labels according to a list to construct a photovoltaic module list;

[0081] Traversing the photovoltaic component list, obtaining a first photovoltaic component, retrieving design parameters of the first photovoltaic component, and extracting a first parameter range from the design parameters according to a preset parameter dimension;

[0082] The first photovoltaic component is associated with the first parameter range and stored, and added to the photovoltaic range data set.

[0083] Furthermore, the simulation platform building module 13 includes the following execution steps:

[0084] Determine a photovoltaic module topology structure according to the target photovoltaic network, wherein the photovoltaic module topology structure includes a plurality of topology nodes corresponding to a plurality of photovoltaic modules one by one, and each topology node has a formatted photovoltaic module model;

[0085] Traversing a plurality of photovoltaic components in the photovoltaic range data set, determining a first photovoltaic component, extracting a first parameter range of the first photovoltaic component, and determining a first topological node corresponding to the first photovoltaic component in a photovoltaic component topological structure;

[0086] A formatted photovoltaic component model in the first topology node is set based on the first parameter range to obtain a virtual simulation platform.

[0087] Furthermore, the mapping relationship determination module 14 includes the following execution steps:

[0088] Preprocessing the historical power generation data and the historical meteorological data to obtain a plurality of power generation data sequences and a plurality of meteorological data sequences, wherein the plurality of power generation data sequences correspond to the plurality of meteorological data sequences one by one;

[0089] According to multiple power generation data sequences and multiple meteorological data sequences, the meteorological parameters in the meteorological data are locked, the influence on the power generation data is analyzed, and the influence weight of the meteorological parameters is obtained;

[0090] A power generation-meteorological mapping relationship is obtained based on the meteorological parameter impact weights.

[0091] Furthermore, the mapping relationship determination module 14 further includes the following execution steps:

[0092] Traversing and extracting a first power generation data sequence and a first meteorological data sequence from a plurality of power generation data sequences and a plurality of meteorological data sequences;

[0093] In the virtual simulation platform, a first meteorological parameter is determined in the first meteorological data sequence, and other meteorological parameters are locked to obtain a first meteorological impact value of the first meteorological parameter on the first power generation data sequence;

[0094] A first meteorological parameter influence weight is obtained according to the first meteorological influence value, and is added to the meteorological parameter influence weight.

[0095] Furthermore, the sample construction module 15 includes the following execution steps:

[0096] Extracting the meteorological parameter impact weights in the power generation-meteorological mapping relationship, marking the historical meteorological data based on the meteorological parameter impact weights, and acquiring initial meteorological data;

[0097] Meteorological data without weight identification in the initial meteorological data are eliminated to obtain identified meteorological data, and a meteorological training sample set is obtained based on the identified meteorological data.

[0098] Furthermore, the embodiment of the present application also includes a model optimization module, which includes the following execution steps:

[0099] Obtaining the prediction accuracy of the weather forecast information;

[0100] When the prediction accuracy is lower than the accuracy threshold, the meteorological prediction model is optimized.

[0101] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0102] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A refined meteorological forecasting method based on photovoltaic range, characterized in that: The method comprises: Determine a target photovoltaic network, the target photovoltaic network includes a plurality of photovoltaic components, collect parameter ranges of the plurality of photovoltaic components respectively, and construct a photovoltaic range data set; According to the target photovoltaic network, historical power generation data is collected, and corresponding historical meteorological data is simultaneously collected; Building a virtual simulation platform based on the plurality of photovoltaic components and in combination with the photovoltaic range data set; Importing the historical power generation data and the historical meteorological data into the virtual simulation platform, performing correlation analysis between the power generation data and the meteorological data, and determining a power generation-meteorological mapping relationship; Based on the power generation-meteorological mapping relationship and according to the historical meteorological data, a meteorological training sample set is constructed; A meteorological prediction model is constructed using the meteorological training sample set to provide meteorological prediction information for the target photovoltaic network.

2. The method according to claim 1, characterized in that Collect parameter ranges of multiple photovoltaic modules separately to build a photovoltaic range data set, including: Uniquely labeling the plurality of photovoltaic modules respectively, obtaining a plurality of photovoltaic module labels, and storing the plurality of photovoltaic module labels according to a list to construct a photovoltaic module list; Traversing the photovoltaic component list, obtaining a first photovoltaic component, retrieving design parameters of the first photovoltaic component, and extracting a first parameter range from the design parameters according to a preset parameter dimension; The first photovoltaic component is associated with the first parameter range and stored, and added to the photovoltaic range data set.

3. The method according to claim 2, characterized in that Based on the plurality of photovoltaic components and in combination with the photovoltaic range data set, a virtual simulation platform is built, including: Determine a photovoltaic module topology structure according to the target photovoltaic network, wherein the photovoltaic module topology structure includes a plurality of topology nodes corresponding to a plurality of photovoltaic modules one by one, and each topology node has a formatted photovoltaic module model; Traversing a plurality of photovoltaic components in the photovoltaic range data set, determining a first photovoltaic component, extracting a first parameter range of the first photovoltaic component, and determining a first topological node corresponding to the first photovoltaic component in a photovoltaic component topological structure; A formatted photovoltaic component model in the first topology node is set based on the first parameter range to obtain a virtual simulation platform.

4. The method according to claim 1, characterized in that: Importing the historical power generation data and the historical meteorological data into the virtual simulation platform, performing correlation analysis between the power generation data and the meteorological data, and determining the power generation-meteorological mapping relationship, including: Preprocessing the historical power generation data and the historical meteorological data to obtain a plurality of power generation data sequences and a plurality of meteorological data sequences, wherein the plurality of power generation data sequences correspond to the plurality of meteorological data sequences one by one; According to multiple power generation data sequences and multiple meteorological data sequences, the meteorological parameters in the meteorological data are locked, the influence on the power generation data is analyzed, and the influence weight of the meteorological parameters is obtained; A power generation-meteorological mapping relationship is obtained based on the meteorological parameter impact weights.

5. The method according to claim 4, characterized in that Lock the meteorological parameters in the meteorological data, analyze the impact on the power generation data, and obtain the impact weight of the meteorological parameters, including: Traversing and extracting a first power generation data sequence and a first meteorological data sequence from a plurality of power generation data sequences and a plurality of meteorological data sequences; In the virtual simulation platform, a first meteorological parameter is determined in the first meteorological data sequence, and other meteorological parameters are locked to obtain a first meteorological impact value of the first meteorological parameter on the first power generation data sequence; A first meteorological parameter influence weight is obtained according to the first meteorological influence value, and is added to the meteorological parameter influence weight.

6. The method according to claim 1, characterized in that Based on the power generation-meteorological mapping relationship and according to the historical meteorological data, a meteorological training sample set is constructed; Extracting the meteorological parameter impact weights in the power generation-meteorological mapping relationship, marking the historical meteorological data based on the meteorological parameter impact weights, and acquiring initial meteorological data; Meteorological data without weight identification in the initial meteorological data are eliminated to obtain identified meteorological data, and a meteorological training sample set is obtained based on the identified meteorological data.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining the prediction accuracy of the weather forecast information; When the prediction accuracy is lower than the accuracy threshold, the meteorological prediction model is optimized.

8. A refined weather forecast system based on photovoltaic range, characterized in that: The system comprises: A data set construction module, wherein the data set construction module is used to determine a target photovoltaic network, wherein the target photovoltaic network includes a plurality of photovoltaic components, and respectively collect parameter ranges of the plurality of photovoltaic components to construct a photovoltaic range data set; A historical data collection module, the historical data collection module is used to collect historical power generation data according to the target photovoltaic network, and simultaneously collect corresponding historical meteorological data; A simulation platform building module, the simulation platform building module is used to build a virtual simulation platform based on the multiple photovoltaic components and in combination with the photovoltaic range data set; A mapping relationship determination module, the mapping relationship determination module is used to import the historical power generation data and the historical meteorological data into the virtual simulation platform, perform correlation analysis between the power generation data and the meteorological data, and determine the power generation-meteorological mapping relationship; A sample construction module, the sample construction module is used to construct a meteorological training sample set based on the power generation-meteorological mapping relationship and the historical meteorological data; A meteorological prediction module is used to construct a meteorological prediction model using the meteorological training sample set to provide meteorological prediction information for the target photovoltaic network.

Images