Photovoltaic power generation power prediction method and system based on radiation wave fluctuation hierarchical clustering

Abstract

The application discloses a photovoltaic power generation power prediction method and system based on radiation fluctuation hierarchical clustering, comprising: preprocessing power generation power data and meteorological data of the photovoltaic power generation system to obtain preprocessed power generation power data and meteorological data; calculating radiation intensity features and radiation fluctuation features based on the preprocessed meteorological data; performing first layer clustering processing on the preprocessed weather data based on the radiation intensity features, dividing the weather into a plurality of initial categories to obtain a first layer clustering result; performing subdivided clustering on the result of the first layer clustering based on the radiation fluctuation features to obtain hierarchical weather types; obtaining key meteorological features; inputting the key meteorological features and the hierarchical weather types into a trained photovoltaic power generation power prediction model to predict photovoltaic power generation power, and the method and system can improve the accuracy and adaptability of photovoltaic power generation power prediction under complex meteorological conditions.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic power generation prediction and meteorological data analysis technology, and relates to a photovoltaic power generation prediction method and system based on hierarchical clustering of radiation fluctuations. Background Technology

[0002] As a typical clean energy generation method, photovoltaic (PV) power generation is highly sensitive to external meteorological conditions, especially directly affected by changes in radiation intensity. In actual operation, influenced by factors such as cloud cover, rainfall, and changes in atmospheric conditions, PV power generation often exhibits significant fluctuations and uncertainties. This complex dynamic characteristic not only increases the difficulty of power system dispatch and absorption but also places higher demands on the accuracy and stability of PV power generation prediction models.

[0003] In existing research on photovoltaic (PV) power generation prediction, some methods primarily rely on the mean or threshold of radiation intensity to simply classify weather conditions, using a single radiation index to distinguish weather states under different illumination conditions to reflect their impact on PV power generation. However, these methods are mainly based on static statistical characteristics and struggle to fully characterize the dynamic changes in radiation over time, especially failing to adequately consider the weather instability reflected by radiation fluctuations, thus limiting the ability of classification results to represent complex weather scenarios. Meanwhile, some studies introduce multiple meteorological factors (such as radiation, temperature, and humidity) for comprehensive modeling and classify weather or scenarios based on multidimensional meteorological features. However, these methods have certain limitations in practical applications: on the one hand, the physical interaction mechanisms between different meteorological parameters and their contribution to PV power generation vary, making it difficult to achieve unified quantification and reasonable weight allocation for each parameter in the model; on the other hand, the introduction of multidimensional features significantly increases data dimensionality and model complexity, which can lead to model learning difficulties when the sample size is limited or meteorological data fluctuates greatly, thus affecting classification performance and prediction stability. In machine learning prediction models, the model essentially establishes a mapping relationship between input and output by learning statistical patterns from historical data. However, for data such as photovoltaic power generation that is driven by weather and has significant non-stationary characteristics, when the data simultaneously contains multiple weather scenarios with large differences in intensity and significant fluctuations, unified modeling often fails to effectively capture its inherent patterns, leading to increased model learning difficulty and rising prediction errors. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a photovoltaic power generation prediction method and system based on hierarchical clustering of radiation fluctuations. This method and system can improve the accuracy and adaptability of photovoltaic power generation prediction under complex meteorological conditions.

[0005] To achieve the above objectives, this invention discloses a photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations, comprising: Acquire power generation data and meteorological data from photovoltaic power generation systems; The power generation data and meteorological data of the photovoltaic power generation system are preprocessed to obtain preprocessed power generation data and meteorological data. Based on the preprocessed meteorological data, the radiation intensity characteristics and radiation fluctuation characteristics are calculated. Based on the radiation intensity characteristics, the preprocessed weather data is subjected to a first-level clustering process to divide the weather into several initial categories, thus obtaining the first-level clustering result. Based on the characteristics of radiation fluctuations, the results of the first-level clustering are further subdivided into hierarchical weather types; Based on the preprocessed meteorological data and photovoltaic power generation data, key meteorological features were screened to obtain key meteorological features. The key meteorological features and stratified weather types are input into the trained photovoltaic power generation prediction model to predict the photovoltaic power generation.

[0006] Furthermore, the meteorological data includes total solar radiation, diffuse solar radiation, temperature, wind speed, wind direction, and humidity data.

[0007] Furthermore, the process of preprocessing the power generation data and meteorological data of the photovoltaic power generation system is as follows: The power generation data and meteorological data of the photovoltaic power generation system are processed for missing data, outlier detection and correction.

[0008] Furthermore, the average and maximum values ​​of the daily total radiation are calculated and used as radiation intensity characteristics. By calculating the absolute value of the difference between radiation values ​​at adjacent times within the same day, and averaging the differences of all time intervals, the radiation fluctuation characteristics are obtained. These radiation fluctuation characteristics characterize the severity of radiation changes.

[0009] Furthermore, the K-means clustering algorithm is used to perform the first layer of clustering on the preprocessed weather data based on the radiation intensity characteristics.

[0010] Furthermore, the Pearson correlation coefficient method was used to screen key features of meteorological factors based on preprocessed meteorological data and photovoltaic power generation data.

[0011] Furthermore, the photovoltaic power generation prediction model is trained using a bidirectional gated recurrent unit prediction model based on an attention mechanism.

[0012] This invention discloses a photovoltaic power generation prediction system based on hierarchical clustering of radiation fluctuations, comprising: The acquisition module is used to acquire power generation data and meteorological data of the photovoltaic power generation system; The preprocessing module is used to preprocess the power generation data and meteorological data of the photovoltaic power generation system to obtain preprocessed power generation data and meteorological data. The calculation module is used to calculate the radiation intensity characteristics and radiation fluctuation characteristics based on the preprocessed meteorological data; The first clustering module is used to perform a first-level clustering process on the preprocessed weather data based on the radiation intensity characteristics, dividing the weather into several initial categories to obtain the first-level clustering result; The second clustering module is used to further subdivide the results of the first-level clustering based on the radiation fluctuation characteristics to obtain hierarchical weather types. The filtering module is used to filter key features of meteorological factors based on preprocessed meteorological data and photovoltaic power generation data to obtain key meteorological features. The prediction module is used to input the key meteorological features and stratified weather types into the trained photovoltaic power generation prediction model to predict the photovoltaic power generation.

[0013] This invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations.

[0014] The present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the photovoltaic power generation prediction method based on radiation fluctuation hierarchical clustering.

[0015] The present invention has the following beneficial effects: The photovoltaic power generation prediction method and system based on radiation fluctuation hierarchical clustering described in this invention, in specific operation, takes radiation data as the basis, constructs radiation intensity characteristics and fluctuation characteristics, and performs hierarchical characterization and classification of weather conditions, thereby achieving effective differentiation of different weather types. On this basis, a photovoltaic power generation prediction model is introduced to predict photovoltaic power generation, thereby improving the prediction accuracy and model adaptability under complex meteorological conditions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a comparison chart of photovoltaic power generation characteristics under various weather conditions based on the classification of radiation fluctuation characteristics according to the present invention; Figure 3 The graph shows the predicted photovoltaic power generation under various weather conditions according to the classification of this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0020] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0022] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0023] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0026] Example 1 refer to Figure 1 The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations described in this invention includes the following steps: 1) Obtain historical power generation data and meteorological data of photovoltaic power generation systems; The meteorological data includes total solar radiation, diffuse solar radiation, temperature, wind speed, wind direction, and humidity data. The power generation data and meteorological data have a time series resolution of minutes or hours, which are used for subsequent weather feature analysis and photovoltaic power generation prediction modeling.

[0027] 2) Preprocess the historical photovoltaic power generation data and meteorological data obtained; Historical photovoltaic power generation data and meteorological data undergo missing data processing, outlier detection, and correction. For missing data processing, data integrity is assessed on a daily basis. If any data point within a day is missing, the data for that day is considered incomplete, and all data for that day is removed to ensure the integrity and continuity of the time series data. For outlier detection and correction, the Pauta Criterion method based on statistical characteristics is used. By calculating the mean and standard deviation of the data, data points deviating from the overall distribution range are identified as outliers. Detected outliers are then removed or corrected to reduce their impact on model training.

[0028] 3) Based on the preprocessed meteorological data, statistical analysis is performed on the radiation data on a daily basis to calculate the average and maximum daily total radiation. This average and maximum daily total radiation are used to characterize the overall radiation intensity level under different weather conditions; that is, the average and maximum daily total radiation are used as radiation intensity features. Simultaneously, to depict the variation characteristics of radiation over time, a radiation fluctuation feature is introduced. This feature is obtained by calculating the absolute value of the difference between radiation values ​​at adjacent times within the same day, summing the differences over all time intervals, and then averaging the sum, thus characterizing the drastic nature of radiation changes. These radiation intensity and fluctuation features together construct a basic feature set for weather classification.

[0029] The maximum solar radiation is:

[0030] in, This represents the maximum daily radiation. For the first The radiation value at any given time.

[0031] The average daily radiation is:

[0032] in, This represents the average daily radiation. This represents the total amount of radiation data for that day.

[0033] The solar radiation fluctuation is:

[0034] in, This represents the value of solar radiation fluctuation.

[0035] 4) Based on the radiation intensity characteristics, the weather data is subjected to the first-level clustering process to divide the weather into different initial categories to characterize the weather state under different radiation levels, and the first-level clustering results are obtained.

[0036] When using the K-means clustering algorithm (K-means) for clustering, the original data is divided into weather categories with different radiation intensities based on the maximum and average daily radiation values.

[0037] The mathematical model for K-means is:

[0038] in, Indicates the first One cluster; Indicates the first One sample point; Indicates the first Cluster centers; Indicates the number of clusters; Represents the square of the Euclidean distance.

[0039] 5) Based on the first-level clustering results, the data of each category are further subdivided into clusters based on the radiation fluctuation characteristics to obtain several weather types with different fluctuation characteristics, thereby achieving a hierarchical division of weather conditions.

[0040] 6) Based on the preprocessed historical meteorological data and photovoltaic power generation data obtained in step 2), the Pearson correlation coefficient method is used to screen key features of meteorological factors and extract the key meteorological features that have the most significant impact on photovoltaic power generation, so as to provide effective input for subsequent prediction models and improve model training efficiency and prediction accuracy.

[0041] The Pearson correlation coefficient is:

[0042] in, Represents the correlation coefficient; Indicates the first Sample values ​​of meteorological factors; This represents the sample value of photovoltaic power generation. and These represent the mean values ​​of the corresponding sequences; Indicates the number of samples.

[0043] 7) Using the key meteorological features selected in step 6) and several weather types with different fluctuation characteristics obtained in step 5), construct attention-based bidirectional gated recurrent unit (Attention-BiGRU) prediction models. Use the meteorological features under different weather types as model input, and perform bidirectional GRU feature extraction on the input sequences. The hidden state is represented as follows:

[0044] in, Indicates the first The hidden state of the bidirectional GRU output at any given time; This indicates the hidden state of the forward GRU; This indicates the backward GRU hidden state.

[0045] Based on this, an attention mechanism is introduced to assign weights to features at different time steps. The attention weights are:

[0046] in, Indicates the first Attention weights at each moment; Indicates attention score; Indicates the length of the time series; It is an exponential function.

[0047] The final photovoltaic power prediction result obtained by weighted summation is as follows:

[0048] Example 2 This embodiment takes a photovoltaic power station in Kunming, Yunnan Province as an example and includes the following steps: 1) Obtain historical power generation data of a photovoltaic power station in Kunming, Yunnan Province. The data includes temperature, total solar radiation, diffuse solar radiation, wind speed, wind direction, and air pressure, with a resolution of 1 hour.

[0049] 2) Preprocessing of the acquired raw data, including missing data handling and outlier detection and correction. Missing data for the day is removed by directly discarding all data from that day. The Laida method is used to detect and process outliers in the raw data. The scope of data processing includes all raw meteorological data and power generation forecast data.

[0050] 3) Based on the preprocessed meteorological data, statistical analysis of radiation data is performed on a daily basis to calculate the average and maximum daily total radiation, which characterizes the overall radiation intensity level under different weather conditions. Hourly calculations are performed on the radiation data for each day to further calculate the radiation fluctuation index. This involves averaging the absolute differences in radiation values ​​at adjacent times within the same day to depict the severity of radiation variations. The daily average, maximum, and fluctuation index are used to construct a basic feature set for weather classification, providing input data for subsequent intensity clustering and fluctuation clustering.

[0051] 4) Based on the radiation intensity characteristics obtained in step 3), the weather data is subjected to the first-level clustering process to obtain the first-level clustering results, which divide the weather into different initial categories to characterize the weather state under different radiation levels. The K-means clustering method is used, with the average and maximum values ​​of the daily total radiation as the clustering basis, to divide the data into two weather categories with different radiation intensities. 5) Based on the first-level clustering results, further sub-clustering based on radiation fluctuation characteristics is performed on each category of data to obtain weather types with different fluctuation characteristics, thus achieving hierarchical classification of weather states. The K-means clustering method is used to cluster each category in the first level into two categories, resulting in four weather types, which are then numbered as shown in Table 1. Figure 2 The characteristics of power generation under various weather conditions are shown. By comparing the changes in photovoltaic power generation under different weather types, the impact of radiation fluctuations on power generation can be clearly observed.

[0052] Table 1

[0053] 6) Based on the preprocessed data obtained in step 2), key factors were screened from historical meteorological data. The Pearson correlation coefficient method was used to analyze the correlation between temperature, total solar radiation, diffuse solar radiation, wind speed, wind direction, and air pressure and photovoltaic power generation. The screening results showed that temperature, total solar radiation, and diffuse solar radiation have a significant impact on photovoltaic power generation. Specific correlations are shown in Table 2. Therefore, these three meteorological factors were used as input features for subsequent prediction models.

[0054] Table 2

[0055] 7) Based on the four weather types obtained in step 5 and the key meteorological factors selected in step 6, corresponding photovoltaic power generation prediction models are established. A bidirectional gated recurrent unit (Attention-BiGRU) model with an attention mechanism is used to train and predict data under different weather types, and the photovoltaic power generation prediction results are output. The prediction results are as follows:Figure 3 As shown, the power generation characteristics and forecasting effects of each weather type can be displayed intuitively, thereby verifying the effectiveness of the hierarchical weather classification method in improving the forecasting accuracy of photovoltaic power generation and adapting to complex weather conditions.

[0056] Example 3 The photovoltaic power generation prediction system based on hierarchical clustering of radiation fluctuations described in this invention includes: The acquisition module is used to acquire power generation data and meteorological data of the photovoltaic power generation system; The preprocessing module is used to preprocess the power generation data and meteorological data of the photovoltaic power generation system to obtain preprocessed power generation data and meteorological data. The calculation module is used to calculate the radiation intensity characteristics and radiation fluctuation characteristics based on the preprocessed meteorological data; The first clustering module is used to perform a first-level clustering process on the preprocessed weather data based on the radiation intensity characteristics, dividing the weather into several initial categories to obtain the first-level clustering result; The second clustering module is used to further subdivide the results of the first-level clustering based on the radiation fluctuation characteristics to obtain hierarchical weather types. The filtering module is used to filter key features of meteorological factors based on preprocessed meteorological data and photovoltaic power generation data to obtain key meteorological features. The prediction module is used to input the key meteorological features and stratified weather types into the trained photovoltaic power generation prediction model to predict the photovoltaic power generation.

[0057] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0058] Example 4 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations. For example, the method includes: acquiring power generation data and meteorological data of a photovoltaic power generation system; preprocessing the power generation data and meteorological data of the photovoltaic power generation system to obtain preprocessed power generation data and meteorological data; calculating radiation intensity characteristics and radiation fluctuation characteristics based on the preprocessed meteorological data; performing a first-level clustering process on the preprocessed weather data based on the radiation intensity characteristics to divide the weather into several initial categories, obtaining a first-level clustering result; further subdividing the first-level clustering result based on the radiation fluctuation characteristics to obtain hierarchical weather types; filtering key features of meteorological factors based on the preprocessed meteorological data and photovoltaic power generation data to obtain key meteorological features; and inputting the key meteorological features and hierarchical weather types into a trained photovoltaic power generation prediction model to predict the photovoltaic power generation. The memory may include main memory, such as high-speed random access memory (RAM), or non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which may be an industry-standard architecture bus, a peripheral component interconnection standard bus, or an extended industry-standard architecture bus. The bus can be categorized as an address bus, data bus, or control bus. The memory stores programs; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0059] Example 5 A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations. For example, the method includes: acquiring power generation data and meteorological data of a photovoltaic power generation system; preprocessing the power generation data and meteorological data to obtain preprocessed power generation data and meteorological data; calculating radiation intensity characteristics and radiation fluctuation characteristics based on the preprocessed meteorological data; performing a first-level clustering process on the preprocessed weather data based on the radiation intensity characteristics to divide the weather into several initial categories, obtaining a first-level clustering result; further subdividing the first-level clustering result based on the radiation fluctuation characteristics to obtain hierarchical weather types; filtering key features of meteorological factors based on the preprocessed meteorological data and photovoltaic power generation data to obtain key meteorological features; and inputting the key meteorological features and hierarchical weather types into a trained photovoltaic power generation prediction model to predict the photovoltaic power generation. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0060] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0065] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0066] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations, characterized in that, include: Acquire power generation data and meteorological data from photovoltaic power generation systems; The power generation data and meteorological data of the photovoltaic power generation system are preprocessed to obtain preprocessed power generation data and meteorological data. Based on the preprocessed meteorological data, the radiation intensity characteristics and radiation fluctuation characteristics are calculated. Based on the radiation intensity characteristics, the preprocessed weather data is subjected to a first-level clustering process to divide the weather into several initial categories, thus obtaining the first-level clustering result. Based on the characteristics of radiation fluctuations, the results of the first-level clustering are further subdivided into hierarchical weather types; Based on the preprocessed meteorological data and photovoltaic power generation data, key meteorological features were screened to obtain key meteorological features. The key meteorological features and stratified weather types are input into the trained photovoltaic power generation prediction model to predict the photovoltaic power generation.

2. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, The meteorological data includes total solar radiation, diffuse solar radiation, temperature, wind speed, wind direction, and humidity data.

3. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, The process of preprocessing the power generation data and meteorological data of the photovoltaic power generation system is as follows: The power generation data and meteorological data of the photovoltaic power generation system are processed for missing data, outlier detection and correction.

4. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, Calculate the average and maximum values ​​of daily total radiation, and use the average and maximum values ​​of daily total radiation as radiation intensity characteristics; The radiation fluctuation characteristics are obtained by calculating the absolute value of the difference between radiation values ​​at adjacent times within the same day, accumulating the differences over all time intervals, and then averaging them. These radiation fluctuation characteristics characterize the degree of intensity of radiation changes.

5. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, The K-means clustering algorithm is used to perform the first layer of clustering on the preprocessed weather data based on the radiation intensity characteristics.

6. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, The Pearson correlation coefficient method was used to screen key features of meteorological factors based on preprocessed meteorological data and photovoltaic power generation data.

7. The photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations according to claim 1, characterized in that, The photovoltaic power generation prediction model is trained using a bidirectional gated recurrent unit prediction model based on an attention mechanism.

8. A photovoltaic power generation prediction system based on hierarchical clustering of radiation fluctuations, characterized in that, include: The acquisition module is used to acquire power generation data and meteorological data of the photovoltaic power generation system; The preprocessing module is used to preprocess the power generation data and meteorological data of the photovoltaic power generation system to obtain preprocessed power generation data and meteorological data. The calculation module is used to calculate the radiation intensity characteristics and radiation fluctuation characteristics based on the preprocessed meteorological data; The first clustering module is used to perform a first-level clustering process on the preprocessed weather data based on the radiation intensity characteristics, dividing the weather into several initial categories to obtain the first-level clustering result; The second clustering module is used to further subdivide the results of the first-level clustering based on the radiation fluctuation characteristics to obtain hierarchical weather types. The filtering module is used to filter key features of meteorological factors based on preprocessed meteorological data and photovoltaic power generation data to obtain key meteorological features. The prediction module is used to input the key meteorological features and stratified weather types into the trained photovoltaic power generation prediction model to predict the photovoltaic power generation.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the photovoltaic power generation prediction method based on hierarchical clustering of radiation fluctuations as described in any one of claims 1-7.

Images