Photovoltaic power generation power prediction method, system and device and storage medium

Abstract

The invention provides a photovoltaic power generation power prediction method, system and device and a storage medium, and relates to the technical field of photovoltaic power generation power prediction, and the method comprises the steps: collecting historical data and real-time data of a photovoltaic array region; performing synchronous preprocessing to generate a historical data set and a real-time data set; establishing an initial generation power model, and drawing a curve graph of generation power changing along with time; analyzing climate data and cloud cluster image data in the real-time data, judging the illumination condition of the photovoltaic array area, and calculating the moving direction and the moving speed of the photovoltaic array area under the illumination condition; calculating the actual illuminated area of the photovoltaic array area under each time node, introducing the information into the initial power generation power model for optimization, and generating new prediction data; and carrying out comparison verification on the optimized prediction data and subsequent actual generation power. According to the method, the prediction accuracy is remarkably improved, and the influence of environmental factors on the prediction result is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation prediction, and in particular to a photovoltaic power generation prediction method, system, device and storage medium. Background Art

[0002] With the growing global demand for sustainable energy, new energy power generation technology, especially photovoltaic power generation, has become one of the important ways to solve the energy crisis and environmental pollution. The power generation of photovoltaic power generation systems is directly affected by the intensity of solar radiation. Therefore, accurate prediction of new energy power generation is of great significance for the stable operation of the power system, optimal resource allocation, and effective scheduling of the power market.

[0003] However, current prediction methods for renewable energy power generation, especially in the field of photovoltaic power generation, generally face the challenge of insufficient prediction accuracy. Traditional prediction methods mainly rely on meteorological data, such as solar radiation intensity, temperature, humidity, etc. Although these parameters can reflect the overall status of solar energy resources to a certain extent, their prediction capabilities are limited by many factors in practical applications. Specifically, meteorological data can usually only provide relatively macroscopic solar radiation information, such as whether there is solar radiation, but it is difficult to accurately capture the dynamic changes of clouds in the sky and their specific impact on the light receiving area of ​​the photovoltaic array.

[0004] The existence of clouds and their movement paths, shapes, thickness and other characteristics have a direct impact on the intensity of sunlight received by the photovoltaic array. Clouds blocking sunlight will cause a sudden drop in the power generation of the photovoltaic array, and this impact is often sudden and regional. Traditional meteorological data forecasting methods are difficult to accurately reflect such local and rapidly changing lighting conditions in real time. Therefore, the forecast results based on single meteorological data often have a large deviation from the actual power generation, which cannot meet the high-precision requirements of the power system for photovoltaic power generation forecasting.

[0005] In order to improve the accuracy and reliability of photovoltaic power generation prediction, it is necessary to develop a new prediction method that can comprehensively consider the dynamic changes of clouds and their impact range. This method needs to be able to capture detailed information about clouds in the sky in real time, and accurately calculate the area of ​​solar illumination that the photovoltaic array can receive at different times, so as to achieve a refined prediction of photovoltaic power generation. Summary of the invention

[0006] To this end, an embodiment of the present invention provides a photovoltaic power generation power prediction method, system, device and storage medium, which are used to solve the problem that the photovoltaic power generation power prediction technology in the prior art only relies on meteorological data, which is difficult to accurately capture the dynamic changes of clouds and their specific impact on the light receiving area of ​​the photovoltaic array, resulting in deviations in the predicted power data.

[0007] In order to solve the above problem, an embodiment of the present invention provides a photovoltaic power prediction method, which includes:

[0008] Step S1: Collect historical data and real-time data of the photovoltaic array area;

[0009] Step S2: Synchronously preprocess the historical data and real-time data using a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets;

[0010] Step S3: Based on the historical data set, sort the power generation data by time node, establish an initial power generation model, and draw a curve chart of power generation changing over time;

[0011] Step S4: Analyze the climate data and cloud image data in the real-time data to determine the lighting conditions in the photovoltaic array area. Under lighting conditions, process the cloud image to identify cloud features and calculate its moving direction and speed.

[0012] Step S5: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data;

[0013] Step S6: Compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

[0014] Preferably, in step S2, the historical data and the real-time data are synchronously preprocessed by a preset data set template, matching and filling values ​​according to data types, and generating the historical data set and the real-time data set, which specifically includes:

[0015] Step S21: design and construct a data set template with a search and selection function according to the data type required for subsequent data processing and analysis; introduce the collected historical data and real-time data into the constructed data set template;

[0016] Step S22: The data set template matches the type titles of the imported historical data and real-time data according to the data type titles defined within it; once the title text features are successfully matched, the numerical data after the corresponding data type title will be automatically searched and filled into the numerical row and column corresponding to the data set template;

[0017] Step S23: After the matching and filling operations in step S22, the values ​​in the data set template will be gradually filled; when all the values ​​in the template are matched, a standardized historical data set and real-time data set are formed.

[0018] Preferably, in step S3, based on the historical data set, the power generation data is sorted by time node, an initial power generation model is established, and a curve chart of power generation changing with time is drawn, which specifically includes:

[0019] Step S31: extracting power generation values ​​from the historical data set and sorting them according to the corresponding time nodes; based on the sorted historical data, using statistical analysis or machine learning algorithms, establishing an initial power generation model;

[0020] Step S32: using the initial power generation model, generate the corresponding relationship between the time node and the actual power generation, and draw it into a curve image; according to the periodic characteristics reflected by the curve image and the corresponding time period, predict the power generation of the subsequent time period.

[0021] Preferably, in step S4, the climate data and cloud image data in the real-time data are analyzed to determine the illumination conditions of the photovoltaic array area. Under illumination conditions, the cloud image is processed to identify cloud features and calculate its moving direction and moving speed, which specifically includes:

[0022] Step S41: Determine whether there is light in the photovoltaic array area according to the climate data in the prediction time period; if there is no light in the entire prediction period, directly predict the no-light power in the initial power generation model; if there is light at some time nodes, execute step S42;

[0023] Step S42: During the time period when the light is confirmed to be present, the cloud image data above the photovoltaic array area is processed as follows:

[0024] Sorting and feature comparison: starting from the real-time time node, sort the cloud image data above the photovoltaic array area, and perform feature comparison on adjacent cloud image data in turn;

[0025] Grayscale and feature labeling: grayscale the cloud image data, and introduce historical feature data to label the corresponding features in the cloud image data, where the cloud feature is labeled as M;

[0026] Calculation of moving direction and moving speed: In the plane image, the X-axis and Y-axis coordinate systems are established based on the image edge, and the lower left corner of the cloud image data is taken as the center point. The coordinates of the center of the cloud feature M in the previous cloud image data are (x1, y1), and the coordinates of the center of the cloud feature M' in the next cloud image data are (x2, y2). By comparing the values ​​of x1 and x2, y1 and y2, the moving direction of the cloud feature is determined, and the moving speed V of the cloud feature is calculated.

[0027] Preferably, the method for calculating the moving speed V of the cloud characteristics is:

[0028]

[0029] Where V represents the moving speed of the cloud characteristics, It is expressed as the distance moved from cloud feature M to cloud feature M' at time t1.

[0030] Preferably, in step S5, according to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data, specifically including:

[0031] Step S51: According to the initial power generation model, the area data of real-time cloud characteristics is introduced, and the shielding area located on the ground is obtained in combination with historical data;

[0032] Step S52: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, which specifically includes:

[0033] An angle value α is generated between the moving direction of the cloud feature and the boundary of the photovoltaic array area, and the cloud feature moves at the angle value α and the moving speed V;

[0034] According to the moving direction of the cloud cluster feature, determine whether the left and right boundaries of the ground shielding of the cloud cluster feature will shield the photovoltaic array area, and calculate the area of ​​the photovoltaic array area that can receive light according to the shielding situation:

[0035] If the left and right boundaries of the ground shielding of the cloud feature shield the boundary of the photovoltaic array area, the shielded area is determined according to the intersection of the front boundary of the cloud feature moving direction and the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated;

[0036] If the left and right side boundaries of the ground shielding of the cloud feature shield one side boundary of the photovoltaic array area, the shielded area is determined according to the boundary of the cloud feature located on the photovoltaic array area and the intersection of the front boundary and the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated;

[0037] If the left and right side boundaries of the ground shielding of the cloud feature do not shield the boundary of the photovoltaic array area, the shielded area is determined according to the intersection of the left and right side boundaries and the front side boundary of the cloud feature with the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated;

[0038] Step S53: introducing the calculated photovoltaic array area that can receive light into the initial power generation model, combining the solar radiation intensity, ambient temperature and the conversion efficiency of the photovoltaic array, and calculating the new power prediction data P at the corresponding time node.

[0039] Preferably, the power prediction data P is calculated as follows:

[0040] P = S*n*I*T;

[0041] Among them, S is the area of ​​the photovoltaic array that can receive light, n is the conversion efficiency of the photovoltaic array, I is the solar radiation intensity, and T is the ambient temperature.

[0042] The embodiment of the present invention further provides a photovoltaic power generation prediction system, which is used to implement the photovoltaic power generation prediction method described above, and specifically includes:

[0043] Data acquisition module, used to collect historical data and real-time data of the photovoltaic array area;

[0044] The data preprocessing module is used to synchronously preprocess historical data and real-time data through a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets;

[0045] The initial power generation model building module is used to sort the power generation data by time nodes based on the historical data set, build the initial power generation model, and draw a curve chart of the power generation changing over time;

[0046] The illumination condition and cloud characteristics analysis module is used to analyze the climate data and cloud images in the real-time data, determine the illumination condition of the photovoltaic array area, and process the cloud images to identify the cloud characteristics and calculate its moving direction and speed under illumination conditions;

[0047] The power generation prediction optimization module is used to calculate the actual area of ​​the photovoltaic array area illuminated at each time node according to the moving direction and moving speed of the cloud characteristics and the location of the photovoltaic array area, and introduce this information into the initial power generation model for optimization to generate new prediction data;

[0048] The prediction data verification module is used to compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

[0049] An embodiment of the present invention also provides an electronic device, which includes a processor, a memory and a bus system, wherein the processor and the memory are connected through the bus system, the memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the photovoltaic power generation prediction method described above.

[0050] An embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores a computer software product, and the computer software product includes a number of instructions for enabling a computer device to execute the photovoltaic power generation power prediction method described above.

[0051] It can be seen from the above technical solutions that the present invention has the following beneficial effects:

[0052] (1) Efficient and accurate prediction: The present invention uses a preset data set template to synchronously process historical and real-time data to establish and optimize the power generation model. It combines historical trends with real-time lighting conditions, especially cloud movement characteristics, to dynamically adjust the prediction data, significantly improving the prediction accuracy while maintaining the efficiency of the prediction process.

[0053] (2) Strong environmental adaptability: The present invention can analyze climate data and cloud image data in real time, accurately calculate the cloud's shielding of the photovoltaic array and its movement trend, and effectively reduce the impact of environmental factors on the prediction results. This prediction method based on real-time meteorological and cloud dynamic changes makes the prediction results closer to the actual power generation situation.

[0054] (3) Intelligent shading analysis: This invention can identify the characteristics of clouds and their moving directions, intelligently determine the shading boundary and area of ​​clouds on the photovoltaic array, and achieve more refined power prediction. Combined with comprehensive analysis of meteorological, cloud maps and ground shading conditions, it greatly improves the accuracy of prediction and simplifies the complexity of prediction operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the implementation cases of the present invention or the technical solutions in the prior art, the following is a brief description of the drawings required for use in the embodiments. By referring to the drawings, the features and advantages of the present invention will be more clearly understood. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0056] Figure 1 A flowchart of a photovoltaic power generation prediction method provided in an embodiment;

[0057] Figure 2 A flowchart for generating a historical data set and a real-time data set in an embodiment;

[0058] Figure 3 A flowchart for establishing an initial power generation model in the embodiment and drawing a curve chart of power generation changing with time;

[0059] Figure 4This is a flow chart of calculating the characteristic moving direction and moving speed of a cloud cluster in an embodiment;

[0060] Figure 5 A flowchart for optimizing the initial power generation model and generating new prediction data in the embodiment;

[0061] Figure 6 It is a block diagram of a photovoltaic power generation prediction system provided in an embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Embodiment 1

[0064] In order to solve the problem that the photovoltaic power generation prediction technology in the existing technology only relies on meteorological data, it is difficult to accurately capture the dynamic changes of clouds and their specific impact on the photovoltaic array light receiving area, resulting in deviations in the predicted power data. Figure 1 As shown, an embodiment of the present invention provides a photovoltaic power generation power prediction method, the method comprising:

[0065] Step S1: Collect historical data and real-time data of the photovoltaic array area;

[0066] Step S2: Synchronously preprocess the historical data and real-time data using a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets;

[0067] Step S3: Based on the historical data set, sort the power generation data by time node, establish an initial power generation model, and draw a curve chart of power generation changing over time;

[0068] Step S4: Analyze the climate data and cloud image data in the real-time data to determine the lighting conditions in the photovoltaic array area. Under lighting conditions, process the cloud image to identify cloud features and calculate its moving direction and speed.

[0069] Step S5: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data;

[0070] Step S6: Compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

[0071] It can be seen from the above technical solutions that the present invention provides a photovoltaic power prediction method, which constructs an initial model by integrating historical data, and combines real-time climate and cloud image data for precise optimization, thereby significantly improving the accuracy of power prediction. This method cleverly combines big data processing and image processing technology to ensure that the prediction model is adjusted in real time based on historical trends to reflect the impact of cloud dynamics on lighting conditions. In particular, the precise analysis of the direction and rate of cloud movement enables the prediction to dynamically reflect the actual situation of the photovoltaic array being blocked, thereby effectively reducing the prediction error caused by environmental factors. This system not only improves the accuracy of the predicted values, but also maintains efficient processing capabilities, achieves a close fit between the prediction results and the actual power generation situation, and provides more reliable technical support for photovoltaic power generation scheduling and management. Its innovative data processing and model optimization strategies have opened up new paths for accurate prediction in the field of photovoltaic power generation.

[0072] In step S1, the present invention collects historical data and real-time data of the photovoltaic array area. Data is obtained from various sensors and data sources at regular intervals to ensure the integrity and accuracy of the data. These data include but are not limited to power generation, climate data (such as temperature, humidity, wind speed, wind direction, etc.), solar radiation intensity, cloud images, etc.

[0073] In step S2, the present invention performs synchronous preprocessing on historical data and real-time data through a preset data set template, matches and fills in values ​​according to data types, and generates historical data sets and real-time data sets, such as Figure 2 As shown, specifically including:

[0074] Step S21: Constructing a dataset template: First, design and construct a dataset template with a search and selection function based on the data type required for subsequent data processing and analysis. This template will be used to uniformly and standardizedly process data from different sources to ensure the consistency and comparability of data types. Subsequently, the received historical data and real-time data are introduced into the constructed dataset template to prepare for subsequent data processing.

[0075] Step S22: Data type matching and value filling: The data set template will match the type titles of the imported historical data and real-time data according to the data type titles defined within it. The key to this step is to ensure that the type titles in the data set template are consistent with the type titles of the actual data in terms of text features. Once the title text features are matched successfully, the numerical data after the corresponding data type title will be automatically searched and filled into the corresponding numerical rows and columns of the data set template. This operation ensures the accuracy and standardization of the data, and provides a solid foundation for subsequent data processing and analysis.

[0076] Step S23: Complete data set generation: After the above matching and filling operations, the values ​​in the data set template will be gradually filled. When all the values ​​in the template are matched, a standardized historical data set and a real-time data set are formed. These two data sets will serve as the basis for subsequent data analysis, model building and prediction, providing strong data support for the accurate prediction of photovoltaic power generation.

[0077] In summary, step S2 synchronously preprocesses historical data and real-time data through a preset data set template, which not only improves the efficiency and accuracy of data processing, but also ensures the consistency and standardization of the data, laying a solid foundation for subsequent power generation forecasting work.

[0078] In this embodiment, in step S3, the present invention sorts the power generation data by time node based on the historical data set, establishes an initial power generation model, and draws a curve chart of power generation changing with time, such as Figure 3 As shown, specifically including:

[0079] Step S31: Data sorting and model initialization: First, extract the power generation values ​​from the historical data set and sort them strictly according to the corresponding time nodes. This is to ensure that the model can accurately reflect the impact of time on power generation. Then, based on the sorted historical data, use statistical analysis or machine learning algorithms to establish an initial power generation prediction model. This model will serve as the basic framework for subsequent predictions and is used to preliminarily predict the power generation at future time nodes.

[0080] Step S32: Curve drawing and period prediction: Subsequently, the initial power generation model is used to generate the corresponding relationship between the time node and the actual power generation, and the relationship is drawn into a curve image. The curve graph intuitively shows the trend of power generation over time, which helps to analyze the changing law of power generation. On this basis, according to the periodic characteristics reflected by the curve image, combined with the corresponding time period (such as day, week, month, etc.), the power generation of the subsequent time period is predicted. The prediction results will serve as a benchmark for subsequent optimization and verification.

[0081] Through the above steps, step S3 not only completes the construction of the initial power generation model, but also intuitively displays the changing trend of the power generation through a curve chart, providing a solid foundation for subsequent optimization and prediction.

[0082] In step S4, the present invention analyzes the climate data and cloud image data in the real-time data to determine the illumination conditions in the photovoltaic array area. Under illumination conditions, the cloud image is processed to identify cloud features (such as shape, size, density, etc.) and calculate its moving direction and moving speed. Figure 4 As shown, specifically including:

[0083] Step S41: Determine the lighting condition: Determine whether there is lighting in the photovoltaic array area based on the climate data within the prediction time period; if there is no lighting during the entire prediction period, directly predict the no-light power in the initial power generation model; if there is lighting at some time nodes, execute step S42.

[0084] Step S42: Cloud image analysis and motion calculation: During the period of time when the light is confirmed to be present, the cloud image data above the photovoltaic array area is processed as follows:

[0085] Sorting and feature comparison: Starting from the real-time time node, the cloud image data located above the photovoltaic array area is sorted, and the feature comparison of adjacent cloud image data is performed in turn.

[0086] Grayscale and feature labeling: The cloud image data is grayscaled, and historical feature data is introduced to label the corresponding features in the cloud image data, where the cloud feature is labeled as M.

[0087] Calculation of moving direction and moving speed: In the plane image, the X-axis and Y-axis coordinate systems are established based on the image edge, and the lower left corner of the cloud image data is taken as the center point. The coordinates of the center of the cloud feature M in the previous cloud image data are (x1, y1), and the coordinates of the center of the cloud feature M' in the next cloud image data are (x2, y2). By comparing the values ​​of x1 and x2, y1 and y2, the moving direction of the cloud feature is determined, and the moving speed V of the cloud feature is calculated.

[0088] The method for calculating the moving speed V of the cloud characteristics is:

[0089]

[0090] Where V represents the moving speed of the cloud cluster characteristics, It is expressed as the distance moved from cloud feature M to cloud feature M' at time t1.

[0091] Through the above steps, step S4 accurately determines the lighting conditions of the photovoltaic array area through detailed climate data analysis and cloud image data processing, and further identifies the characteristics of the cloud and its moving direction and speed under the lighting conditions, providing key data support for subsequent power prediction. This process not only improves the accuracy of the prediction, but also ensures the efficiency and real-time nature of the prediction operation.

[0092] In step S5, according to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data, such as Figure 5 As shown, specifically including:

[0093] Step S51: According to the initial power generation model, the area data of real-time cloud characteristics is introduced, and the shielding area located on the ground is obtained in combination with historical data.

[0094] Step S52: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, which specifically includes:

[0095] An angle value α is generated between the moving direction of the cloud feature and the boundary of the photovoltaic array area, and the cloud feature moves at the angle value α and the moving speed V.

[0096] According to the moving direction of the cloud cluster feature, determine whether the left and right boundaries of the ground shielding of the cloud cluster feature will shield the photovoltaic array area, and calculate the area of ​​the photovoltaic array area that can receive light according to the shielding situation:

[0097] If the left and right boundaries of the ground shielding of the cloud feature block the boundary of the photovoltaic array area, the blocked area is determined according to the intersection of the front boundary of the cloud feature moving direction and the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated.

[0098] If the left and right side boundaries of the ground shielding of the cloud feature shield one side boundary of the photovoltaic array area, the shielded area is determined according to the boundary of the cloud feature located on the photovoltaic array area and the intersection of the front boundary and the photovoltaic array area boundary, and the area of ​​the photovoltaic array area that can receive light is calculated.

[0099] If the left and right boundaries of the ground shielding of the cloud feature do not shield the boundary of the photovoltaic array area, the shielded area is determined according to the intersection of the left and right boundaries of the cloud feature and the front boundary with the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated.

[0100] Step S53: introducing the calculated photovoltaic array area that can receive light into the initial power generation model, combining the solar radiation intensity, ambient temperature and the conversion efficiency of the photovoltaic array, and calculating the new power prediction data P at the corresponding time node.

[0101] The calculation method of the power prediction data P is:

[0102] P = S*n*I*T;

[0103] Where S is the area of ​​the photovoltaic array that can receive light, n is the conversion efficiency of the photovoltaic array, I is the solar radiation intensity, and T is the ambient temperature.

[0104] In step S6, the optimized prediction data is compared and verified with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation and to evaluate the prediction accuracy and reliability.

[0105] In order to evaluate the accuracy of the photovoltaic power prediction method proposed in this invention, a power prediction for the same photovoltaic array area was conducted for a week (divided into multiple time nodes, each node was 30 minutes apart, and a total of 336 time periods). The experimental results show that:

[0106] Using the existing photovoltaic power prediction method, among the predictions of 336 time periods (30 minutes each), only the prediction results of 263 time periods are consistent with the actual power generation.

[0107] In contrast, using the photovoltaic power prediction method of the present invention, among the prediction results within the same time period, the predictions in up to 329 time periods matched the actual power generation.

[0108] It can be seen that the photovoltaic power generation power prediction method of the present invention performs well in improving prediction accuracy and is more suitable for power prediction scenarios in photovoltaic array areas.

[0109] Embodiment 2

[0110] like Figure 6 As shown, the present invention provides a photovoltaic power generation prediction system, which is used to implement the photovoltaic power generation prediction method of the above embodiment 1, and specifically includes:

[0111] The data acquisition module 100 is used to collect historical data and real-time data of the photovoltaic array area;

[0112] The data preprocessing module 200 is used to synchronously preprocess the historical data and the real-time data through a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets;

[0113] The initial power generation model building module 300 is used to sort the power generation data by time nodes based on the historical data set, build an initial power generation model, and draw a curve chart of the power generation changing over time;

[0114] The illumination condition and cloud characteristics analysis module 400 is used to analyze the climate data and cloud images in the real-time data, determine the illumination condition of the photovoltaic array area, and process the cloud images to identify the cloud characteristics and calculate the moving direction and moving speed under illumination conditions;

[0115] The power generation prediction optimization module 500 is used to calculate the actual area of ​​the photovoltaic array area illuminated at each time node according to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, and introduce this information into the initial power generation model for optimization to generate new prediction data;

[0116] The prediction data verification module 600 is used to compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

[0117] A photovoltaic power generation prediction system of the present embodiment is used to implement the aforementioned photovoltaic power generation prediction method. Therefore, the specific implementation method of the photovoltaic power generation prediction system can be seen in the implementation method part of the aforementioned photovoltaic power generation prediction method. For example, the data acquisition module 100, the data preprocessing module 200, the initial power generation model establishment module 300, the illumination condition and cloud characteristic analysis module 400, the power generation prediction optimization module 500, and the prediction data verification module 600 are respectively used to implement steps S1, S2, S3, S4, S5, and S6 in the aforementioned photovoltaic power generation prediction method. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part. In order to avoid redundancy, it will not be repeated here.

[0118] Embodiment 3

[0119] An embodiment of the present invention provides an electronic device, which includes a processor, a memory and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the above-mentioned photovoltaic power generation prediction method.

[0120] Embodiment 4

[0121] An embodiment of the present invention provides a computer storage medium, which stores a computer software product. The computer software product includes several instructions for enabling a computer device to execute the above-mentioned photovoltaic power generation prediction method.

[0122] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0125] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A photovoltaic power generation power prediction method, characterized in that: include: Step S1: Collect historical data and real-time data of the photovoltaic array area; Step S2: Synchronously preprocess the historical data and real-time data using a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets; Step S3: Based on the historical data set, sort the power generation data by time node, establish an initial power generation model, and draw a curve chart of power generation changing over time; Step S4: Analyze the climate data and cloud image data in the real-time data to determine the lighting conditions in the photovoltaic array area. Under lighting conditions, process the cloud image to identify cloud features and calculate its moving direction and speed. Step S5: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data; Step S6: Compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

2. The photovoltaic power generation prediction method according to claim 1, characterized in that: In step S2, the historical data and the real-time data are synchronously preprocessed by using a preset data set template, matching and filling values ​​according to data types, and generating historical data sets and real-time data sets, which specifically includes: Step S21: design and construct a data set template with a search and selection function according to the data type required for subsequent data processing and analysis; introduce the collected historical data and real-time data into the constructed data set template; Step S22: The data set template matches the type titles of the imported historical data and real-time data according to the data type titles defined within it; once the title text features are successfully matched, the numerical data after the corresponding data type title will be automatically searched and filled into the numerical row and column corresponding to the data set template; Step S23: After the matching and filling operations in step S22, the values ​​in the data set template will be gradually filled; when all the values ​​in the template are matched, a standardized historical data set and real-time data set are formed.

3. The photovoltaic power generation prediction method according to claim 1, characterized in that: In step S3, based on the historical data set, the power generation data is sorted by time node, an initial power generation model is established, and a curve chart of power generation changing over time is drawn, which specifically includes: Step S31: extracting power generation values ​​from the historical data set and sorting them according to the corresponding time nodes; based on the sorted historical data, using statistical analysis or machine learning algorithms, establishing an initial power generation model; Step S32: using the initial power generation model, generate the corresponding relationship between the time node and the actual power generation, and draw it into a curve image; according to the periodic characteristics reflected by the curve image and the corresponding time period, predict the power generation of the subsequent time period.

4. The photovoltaic power generation prediction method according to claim 1, characterized in that: In step S4, the climate data and cloud image data in the real-time data are analyzed to determine the illumination conditions in the photovoltaic array area. Under illumination conditions, the cloud image is processed to identify cloud features and calculate its moving direction and moving speed, which specifically includes: Step S41: Determine whether there is light in the photovoltaic array area according to the climate data in the prediction time period; if there is no light in the entire prediction period, directly predict the no-light power in the initial power generation model; if there is light at some time nodes, execute step S42; Step S42: During the time period when the light is confirmed to be present, the cloud image data above the photovoltaic array area is processed as follows: Sorting and feature comparison: Starting from the real-time time node, the cloud image data located above the photovoltaic array area is sorted, and the feature comparison of adjacent cloud image data is performed in turn; Grayscale and feature labeling: grayscale the cloud image data, and introduce historical feature data to label the corresponding features in the cloud image data, where the cloud feature is labeled as M; Calculation of moving direction and moving speed: In the plane image, the X-axis and Y-axis coordinate systems are established based on the image edge, and the lower left corner of the cloud image data is taken as the center point. The coordinates of the center of the cloud feature M in the previous cloud image data are (x1, y1), and the coordinates of the center of the cloud feature M' in the next cloud image data are (x2, y2). By comparing the values ​​of x1 and x2, y1 and y2, the moving direction of the cloud feature is determined, and the moving speed V of the cloud feature is calculated.

5. The photovoltaic power generation prediction method according to claim 4, characterized in that: The method for calculating the moving speed V of the cloud characteristics is: Where V represents the moving speed of the cloud characteristics, It is expressed as the distance moved from cloud feature M to cloud feature M' at time t1.

6. The photovoltaic power generation prediction method according to claim 1, characterized in that: In step S5, according to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, and this information is introduced into the initial power generation model for optimization to generate new prediction data, specifically including: Step S51: According to the initial power generation model, the area data of real-time cloud characteristics is introduced, and the shielding area located on the ground is obtained in combination with historical data; Step S52: According to the moving direction and moving speed of the cloud characteristics and the position of the photovoltaic array area, the actual area of ​​the photovoltaic array area illuminated at each time node is calculated, which specifically includes: An angle value α is generated between the moving direction of the cloud feature and the boundary of the photovoltaic array area, and the cloud feature moves at the angle value α and the moving speed V; According to the moving direction of the cloud cluster feature, determine whether the left and right boundaries of the ground shielding of the cloud cluster feature will shield the photovoltaic array area, and calculate the area of ​​the photovoltaic array area that can receive light according to the shielding situation: If the left and right boundaries of the ground shielding of the cloud feature shield the boundary of the photovoltaic array area, the shielded area is determined according to the intersection of the front boundary of the cloud feature moving direction and the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated; If the left and right side boundaries of the ground shielding of the cloud feature shield one side boundary of the photovoltaic array area, the shielded area is determined according to the boundary of the cloud feature located on the photovoltaic array area and the intersection of the front boundary and the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated; If the left and right side boundaries of the ground shielding of the cloud feature do not shield the boundary of the photovoltaic array area, the shielded area is determined according to the intersection of the left and right side boundaries and the front side boundary of the cloud feature with the boundary of the photovoltaic array area, and the area of ​​the photovoltaic array area that can receive light is calculated; Step S53: introducing the calculated photovoltaic array area that can receive light into the initial power generation model, combining the solar radiation intensity, ambient temperature and the conversion efficiency of the photovoltaic array, and calculating the new power prediction data P at the corresponding time node.

7. The photovoltaic power generation prediction method according to claim 6, characterized in that: The calculation method of power prediction data P is: P = S*n*I*T; Among them, S is the area of ​​the photovoltaic array that can receive light, n is the conversion efficiency of the photovoltaic array, I is the solar radiation intensity, and T is the ambient temperature.

8. A photovoltaic power generation prediction system, characterized in that: The system is used to implement the photovoltaic power generation power prediction method according to any one of claims 1 to 7, specifically comprising: Data acquisition module, used to collect historical data and real-time data of the photovoltaic array area; The data preprocessing module is used to synchronously preprocess historical data and real-time data through a preset data set template, match and fill in values ​​according to data types, and generate historical data sets and real-time data sets; The initial power generation model building module is used to sort the power generation data by time nodes based on the historical data set, build the initial power generation model, and draw a curve chart of the power generation changing over time; The illumination condition and cloud characteristics analysis module is used to analyze the climate data and cloud images in the real-time data, determine the illumination condition of the photovoltaic array area, and process the cloud images to identify the cloud characteristics and calculate its moving direction and speed under illumination conditions; The power generation prediction optimization module is used to calculate the actual area of ​​the photovoltaic array area illuminated at each time node according to the moving direction and moving speed of the cloud characteristics and the location of the photovoltaic array area, and introduce this information into the initial power generation model for optimization to generate new prediction data; The prediction data verification module is used to compare and verify the optimized prediction data with the subsequent actual power generation to confirm whether the prediction data meets the threshold range of the actual power generation.

9. An electronic device, characterized in that: The electronic device includes a processor, a memory and a bus system, the processor and the memory are connected through the bus system, the memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the photovoltaic power generation prediction method described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer software product, and the computer software product includes a number of instructions for enabling a computer device to execute the photovoltaic power prediction method according to any one of claims 1 to 7.

Images