A Compound Photovoltaic Power Generation Prediction Method Based on Visual Recognition

Abstract

The present invention relates to the technical field of photovoltaic energy, and discloses a composite photovoltaic power generation prediction method based on visual recognition, including: obtaining equipment data, weather forecast data and historical power generation data of a photovoltaic power station; constructing a photovoltaic power generation prediction model to obtain first prediction data of photovoltaic power generation; collecting environmental data of photovoltaic panels before the prediction day of the photovoltaic power station, performing key feature extraction, and calculating to obtain influence correction factor data; constructing a composite prediction model to obtain second prediction data of photovoltaic power generation. The present invention enhances the accuracy of prediction by obtaining multi-source data, makes a preliminary prediction by combining weather forecast and equipment data, collects environmental data of photovoltaic panels through visual recognition technology, extracts key features and calculates influence correction factors, captures the influence of local environmental changes on power generation, combines the preliminary prediction results with the influence correction factors, constructs a composite prediction model, obtains a more accurate final power generation prediction, and improves the accuracy of prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic energy, and in particular to a compound photovoltaic power generation prediction method based on visual recognition. Background Art

[0002] Since photovoltaic power generation can directly convert solar energy into electricity, it has been increasingly applied in the world. However, its energy is usually unstable, and its actual power generation is also unstable in reality. The volatility of photovoltaic power generation has brought great challenges to power grid system dispatching, power balance, load forecasting, and distribution network operation. In order to ensure the stable operation of the power grid, it is of great significance to conduct photovoltaic power prediction.

[0003] Currently, it mainly relies on time series analysis and machine learning to predict the power generation of photovoltaic power stations. Firstly, it is difficult to obtain data. Secondly, weather data is not precise data with fine granularity. In this case, especially when it is sunny but there are occasional local clouds, it will affect the accuracy of photovoltaic power generation data prediction. Thirdly, regarding the cleanliness of the photovoltaic panels themselves, there is currently no way to obtain this data. Except in scenarios with robotic cleaning, it is considered that the cleanliness of the photovoltaic panels is the highest after robotic cleaning, but there may also be cases where the robotic cleaning is not thorough. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title, but such simplifications or omissions shall not be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a compound photovoltaic power generation prediction method based on visual recognition to solve the above problems.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a compound photovoltaic power generation prediction method based on visual recognition, including: obtaining equipment data, weather forecast data, and historical power generation data of a photovoltaic power station;

[0008] Constructing a photovoltaic power generation prediction model, training the photovoltaic power generation prediction model through the historical power generation data, taking the weather forecast data and equipment data of the prediction day as inputs, and obtaining the first prediction data of photovoltaic power generation;

[0009] Collecting the environmental data of the photovoltaic panels before the prediction day of the photovoltaic power station, extracting key features, and calculating to obtain influence correction factor data;

[0010] Based on the first prediction data and the influencing factor data, a composite prediction model is constructed to obtain the second prediction data of photovoltaic power generation as the final predicted power of photovoltaic power generation.

[0011] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: obtaining the weather forecast data and historical power generation data of the photovoltaic power station includes:

[0012] Obtaining device data, where the device data includes the position of the photovoltaic panel, rated power, conversion efficiency, tilt angle, and azimuth angle;

[0013] Obtaining the weather forecast data at the prediction time, where the weather forecast data at the prediction time includes time, weather type, total radiation, direct radiation, diffuse radiation, reflected radiation, temperature, humidity, and cloud cover;

[0014] Obtaining the historical power generation data of the past one to three years, where the historical power generation data includes the actual weather data, device data, and actual power generation power data of the power station.

[0015] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: obtaining the first prediction data of photovoltaic power generation includes:

[0016] Performing data preprocessing on the device data, weather forecast data, and historical power generation data, sorting the historical power generation data in time series, and classifying it according to time periodicity according to the data date;

[0017] Dividing the historical power generation data into different training sets, validation sets, and test sets according to the time period type and weather type;

[0018] Using a long short-term memory network to construct a photovoltaic power generation prediction model, introducing an attention mechanism, training the photovoltaic power generation prediction model through the historical power generation data, and inputting the device data and weather forecast data into the trained photovoltaic power generation prediction model to obtain the first prediction data of the prediction day.

[0019] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: collecting the environmental data before the prediction day of the photovoltaic power station includes:

[0020] Setting a plurality of camera points around the photovoltaic panel, regularly collecting the image data around the photovoltaic panel of the photovoltaic power station, and extracting the environmental data from the image data through visual recognition, where the environmental data includes sky sunlight data, photovoltaic panel sunlight data, photovoltaic panel heavy snow data, and photovoltaic panel dirt data;

[0021] According to the weather type on the prediction date, determine the day closest to the prediction date with the same weather type as the first reference date, and extract the key features of the first reference date, including photovoltaic panel sunlight data and sky sunlight data;

[0022] Take the day before the prediction date as the second reference date, and extract the key features of the second reference date, including photovoltaic panel heavy snow data and photovoltaic panel dirt data.

[0023] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: extracting key features includes:

[0024] The sky sunlight data represents the sky illuminance. Collect the sky images above the photovoltaic panel during each period of the first reference date, divide the sky images into multiple sub-regions, identify the sky illuminance of each sub-region during each period, and calculate the sky illuminance within each period region;

[0025] The photovoltaic panel sunlight data represents the photovoltaic panel light intensity. Collect the first photovoltaic panel image data during each period of the first reference date, convert the first photovoltaic panel image data into a photovoltaic panel grayscale image, segment the photovoltaic panel grayscale image, label each complete photovoltaic panel region, identify the grayscale value of each photovoltaic panel region, calculate the average grayscale value of the photovoltaic panel, and obtain the photovoltaic panel light intensity within each period region;

[0026] The photovoltaic panel heavy snow data represents the heavy snow coverage of the photovoltaic panel. Collect the latest second photovoltaic panel image data of the second reference date, perform image preprocessing and image segmentation on the second photovoltaic panel image data, convert the segmented image into a binary image, identify the snow-covered area, count the number of white pixels in the binary image, and calculate the heavy snow coverage of the photovoltaic panel;

[0027] The photovoltaic panel dirt data represents the dirt degree of the photovoltaic panel. Collect the latest third photovoltaic panel image data of the second reference date, perform image preprocessing and image segmentation on the third photovoltaic panel image data, identify the pixel amounts of different dirt types in the third photovoltaic panel image data, and the dirt types include dust, stains, and damage. Based on the weighted sum of the pixel amounts of dust, stains, and damage and the total pixel amount of the photovoltaic panel, obtain the dirt degree of the photovoltaic panel.

[0028] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: calculating the obtained influence correction factor data includes:

[0029] Compare the sky illuminance in each time period area on the first reference day with the illuminance corresponding to the total radiation in the same time period in the weather forecast on the first reference day. If the average value of the difference between the two is greater than the change threshold, calculate the sky illuminance error through the difference between the two to obtain the first correction factor;

[0030] Compare the light intensity of the photovoltaic panel in each time period area on the first reference day with the sky illuminance in each time period area, and calculate the light intensity error of the photovoltaic panel through the difference between the two to obtain the second correction factor;

[0031] Calculate the first influence factor based on the snow coverage degree of the photovoltaic panel;

[0032] Calculate the second influence factor based on the dirtiness degree of the photovoltaic panel;

[0033] The first correction factor, the second correction factor, the first influence factor, and the second influence factor are combined to generate influence correction factor data.

[0034] As a preferred solution of the composite photovoltaic power generation prediction method based on visual recognition according to the present invention, wherein: obtaining the second prediction data of photovoltaic power generation includes:

[0035] Based on the first prediction data and the influence factor data, construct a composite prediction model to obtain the corrected second prediction data. The composite prediction model is expressed as:

[0036] ;

[0037] Wherein, represents the first prediction data of photovoltaic power generation, represents the first correction factor, represents the second correction factor, represents the first influence factor, represents the second influence factor, represents the dynamic adjustment factor.

[0038] In a second aspect, the present invention provides a composite photovoltaic power generation prediction system based on visual recognition, including:

[0039] An acquisition module for acquiring equipment data, weather forecast data, and historical power generation data of a photovoltaic power station;

[0040] A construction module for constructing a photovoltaic power generation prediction model, training the photovoltaic power generation prediction model through the historical power generation data, and using the weather forecast data and equipment data as inputs to obtain the first prediction data of photovoltaic power generation;

[0041] A data processing module, configured to collect the environmental data of the photovoltaic panels before the predicted day of the photovoltaic power station, extract key features, and calculate and obtain influence correction factor data;

[0042] An output module, configured to construct a composite prediction model based on the first prediction data and the influence factor data, obtain second prediction data of the photovoltaic power generation, and use it as the final predicted power of the photovoltaic power generation.

[0043] In a third aspect, the present invention provides an electronic device, including:

[0044] A memory and a processor;

[0045] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the composite photovoltaic power generation prediction method based on visual recognition are implemented.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the steps of the composite photovoltaic power generation prediction method based on visual recognition are implemented.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: By obtaining multi-source data, the present invention provides a comprehensive data basis for the model, enhances the accuracy of prediction, trains the model with historical power generation data, and combines weather forecasts and equipment data for preliminary prediction. By using visual recognition technology to collect the environmental data of photovoltaic panels, extract key features and calculate influence correction factors, the impact of local environmental changes on power generation is captured. Combining the preliminary prediction results with the influence correction factors to construct a composite prediction model, a more accurate final power generation prediction is obtained, which not only improves the accuracy of prediction, but also enhances the robustness and adaptability of the model, and can better cope with complex and changeable environmental conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a schematic diagram of the overall process of the composite photovoltaic power generation prediction method based on visual recognition according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0051] In the following description, many specific details are set forth to facilitate a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0052] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0053] The present invention is described in detail in conjunction with schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views showing the device structure are enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0054] At the same time, in the description of the present invention, it should be noted that the orientation or positional relationships indicated by terms such as "upper, lower, inner, and outer" are based on the orientation or positional relationships shown in the accompanying drawings. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0055] Unless otherwise clearly defined and limited in the present invention, the terms "installed, connected, and coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, and can also be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] Refer to Figure 1 , for an embodiment of the present invention, a compound photovoltaic power generation prediction method based on visual recognition is provided, including:

[0057] S101, obtain the device data, weather forecast data, and historical power generation data of the photovoltaic power station;

[0058] S102, construct a photovoltaic power generation prediction model, train the photovoltaic power generation prediction model with historical power generation data, use the weather forecast data and device data of the prediction day as inputs, and obtain the first prediction data of photovoltaic power generation;

[0059] S103, collect the environmental data of the photovoltaic panels before the prediction day of the photovoltaic power station, perform key feature extraction, and calculate and obtain the influence correction factor data;

[0060] S104, based on the first prediction data and the influence factor data, construct a composite prediction model, obtain the second prediction data of photovoltaic power generation, and use it as the final predicted power of photovoltaic power generation.

[0061] In a preferred implementation manner of step S101, obtaining the weather forecast data and historical power generation data of the photovoltaic power station includes:

[0062] Obtain the latest device data, where the device data includes the position of the photovoltaic panel, rated power, conversion efficiency, tilt angle, and azimuth angle;

[0063] Obtain the weather forecast data at the prediction time, where the weather forecast data at the prediction time includes time, weather type, total radiation, direct radiation, diffuse radiation, reflected radiation, temperature, humidity, and cloud cover;

[0064] Obtain the historical power generation data for the past one to three years, where the historical power generation data includes the actual weather data, device data, and actual power generation power data of the power station.

[0065] Specifically, the specific parameters in the device data, weather forecast data, and historical power generation data of the photovoltaic power station can be set and obtained according to the actual application scenario, including but not limited to the parameter types in this embodiment. Among them, the weather forecast data can also include wind speed, rainfall, air pressure, etc., the device data can also include the type of photovoltaic panel, inverter characteristics, etc., the weather type can include sunny, cloudy, overcast, rainy, etc., and the historical power generation data ensures that the data covers a variety of weather types and seasonal changes.

[0066] It should be noted that by obtaining the equipment data, weather forecast data, and historical power generation data of the photovoltaic power station, the model can obtain rich input information, providing a solid foundation for subsequent model training and prediction, improving the prediction accuracy. The equipment data reflects the real-time operating status of the photovoltaic system, the weather forecast data provides future environmental conditions, and the historical power generation data contains the actual performance of the system under different conditions. The combination of the three can more accurately capture the dynamic changes of the photovoltaic system, reduce prediction errors. The historical power generation data can help the model learn the impacts of different weather conditions, equipment states, and seasonal changes on power generation, thereby improving the generalization ability of the model and enabling it to make reliable predictions in various environments.

[0067] In a preferred implementation manner of step S102, constructing a photovoltaic power generation prediction model and obtaining the first prediction data of photovoltaic power generation includes:

[0068] Perform data preprocessing on the equipment data, weather forecast data, and historical power generation data. Sort the historical power generation data in time series and classify it according to time periodicity according to the data date;

[0069] Divide the historical power generation data into different training sets, validation sets, and test sets according to the time period type and weather type. Among them, the period type can be a seasonal cycle or a monthly cycle, etc.;

[0070] Use a long short-term memory network to construct a photovoltaic power generation prediction model, introduce an attention mechanism, train the photovoltaic power generation prediction model through the historical power generation data, and input the equipment data and weather forecast data into the trained photovoltaic power generation prediction model to obtain the first prediction data of the prediction day.

[0071] Specifically, data preprocessing includes data cleaning and normalization, removing or filling missing values, outliers, and noise data to ensure the integrity and accuracy of the data, normalizing or standardizing features with different dimensions to ensure the stability of model training, identifying and processing outliers to ensure the accuracy of the data, and constructing the data in the form of a sliding window. Each window contains historical data for a period of time and is used to predict the power generation power at the next time point.

[0072] The model framework of the photovoltaic power generation prediction model includes that multiple LSTM layers can be stacked, and a Dropout layer can be added after each layer to prevent overfitting. Finally, a Dense layer is used to output the prediction value. At the same time, an attention mechanism (Attention Mechanism) is introduced to make the model pay more attention to important time steps or features, further improving the prediction accuracy. The historical power generation data is divided into a training set, a validation set, and a test set according to the ratio of 7:1.5:1.5 to ensure the generalization ability of the model on unseen data. The model is trained using the training set data, and the loss function is monitored on the validation set. When the loss on the validation set no longer decreases, the training is stopped in advance to avoid overfitting. The performance of the model is evaluated using methods such as K-fold cross-validation, and the hyperparameters of the model are adjusted using methods such as grid search, random search, or Bayesian optimization to find the optimal configuration. Appropriate metrics such as mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²) are used to measure the prediction effect of the model. Finally, the model is updated in real time. As new data is continuously updated, the model is retrained regularly to ensure that it is always in the best state.

[0073] It should be noted that by constructing a photovoltaic power generation prediction model using LSTM, the characteristics of time series data are fully considered. LSTM can effectively capture the complex dynamic relationship between weather conditions and the power generation power of the photovoltaic system, thus providing accurate prediction results. Through continuous optimization and improvement, the prediction accuracy of the model can be gradually improved, providing strong support for the efficient operation of photovoltaic power generation.

[0074] In a preferred implementation manner of step S103, the environmental data collected before the prediction day of the photovoltaic power station includes:

[0075] Multiple camera points are set around the photovoltaic panels to regularly collect image data around the photovoltaic panels of the photovoltaic power station, and environmental data is extracted from the image data through visual recognition. The environmental data includes sky sunlight data, photovoltaic panel sunlight data, photovoltaic panel heavy snow data, and photovoltaic panel dirt data;

[0076] According to the weather type of the prediction day, the day closest to the prediction day and with the same weather type is determined as the first reference day, and the key features extracted from the first reference day include photovoltaic panel sunlight data and sky sunlight data;

[0077] The day before the prediction day is used as the second reference day, and the key features extracted from the second reference day include photovoltaic panel heavy snow data and photovoltaic panel dirt data.

[0078] Specifically, different camera devices can be selected according to requirements when collecting different data to obtain accurate data. Near the photovoltaic panel, ensure that the camera can clearly capture the surface of the photovoltaic panel and the surrounding environment. Multiple cameras can be selected to cover a large-area photovoltaic array. According to the type of data collected, the camera should maintain a certain angle with the photovoltaic panel to avoid overexposure or reflection caused by direct light. The camera can capture the photovoltaic panel and the surrounding environment from the side to capture the required data. At the same time, the timing of collecting various data can be set according to the actual situation. In this embodiment, it is collected once every 15 minutes. Among them, when predicting photovoltaic power generation, in addition to referring to historical actual power generation data and correcting the predicted value to reduce errors, it is also necessary to consider the actual environment of the photovoltaic power station and the situation of the photovoltaic panel in the near future. Selecting the photovoltaic panel sunlight data and sky sunlight data of the nearest date under the same weather type can effectively judge the accuracy of photovoltaic power generation prediction. In addition, the covering of the photovoltaic panel can significantly affect the power generation efficiency of the photovoltaic panel. Collecting the latest real snow data and dirt data of the photovoltaic panel on the prediction date can key correct the power generation prediction value.

[0079] It should be noted that in photovoltaic power generation prediction, obtaining data such as the sunlight of the photovoltaic panel, sky sunlight, snow coverage, and dirt conditions through visual recognition technology can significantly improve the prediction accuracy and the maintenance efficiency of the system.

[0080] In a preferred implementation manner of step S103, extracting key features includes:

[0081] The sky sunlight data represents the sky illuminance. Collect the sky images above the photovoltaic panel in the first reference day at each time period, divide the sky image into multiple sub-regions, identify the sky illuminance of each sub-region at each time period, and calculate the sky illuminance within the region at each time period;

[0082] The photovoltaic panel sunlight data represents the photovoltaic panel lightness. Collect the first photovoltaic panel image data in the first reference day at each time period, convert the first photovoltaic panel image into a photovoltaic panel grayscale image, segment the photovoltaic panel grayscale image, label each complete photovoltaic panel region, identify the grayscale value of each photovoltaic panel region, calculate the average grayscale value of the photovoltaic panel, and obtain the photovoltaic panel lightness within the region at each time period;

[0083] The photovoltaic panel snow data represents the photovoltaic panel snow coverage. Collect the latest second photovoltaic panel image data on the second reference day, perform image preprocessing and image segmentation on the second photovoltaic panel image data, convert the segmented image into a binary image, identify the snow-covered area, count the number of white pixels in the binary image, and calculate the photovoltaic panel snow coverage;

[0084] The photovoltaic panel dirt data represents the dirtiness degree of the photovoltaic panel. Collect the latest third photovoltaic panel image data on the second reference day, perform image preprocessing and image segmentation on the third photovoltaic panel image data, identify the pixel amounts of different dirt types in the third photovoltaic panel image data. The dirt types include dust, stains, and damages. Based on the weighted sum of the pixel amounts of dust, stains, and damages and the total pixel amount of the photovoltaic panel, the dirtiness degree of the photovoltaic panel is obtained.

[0085] Specifically, when collecting sky sunlight data, a fisheye lens or an all-sky camera can be used to take images of the entire sky, capturing elements such as the sun, clouds, and blue sky. The all-sky camera can provide a 360-degree field of view to ensure that no important meteorological information is missed. Install the camera in the center or an open area of the photovoltaic power station to ensure that there are no obstructions interfering with the shooting. Use a convolutional neural network (CNN) to identify the cloud types (such as cirrus clouds, cumulus clouds, stratus clouds, etc.) in the image. Different cloud types have different effects on solar radiation. The diffuse radiation amount can be estimated based on the thickness and distribution of the clouds. Divide the total area above the photovoltaic power station into multiple sub-regions. By analyzing the color and brightness distribution in the image and combining with the atmospheric transparency model, obtain the light intensity in each direction of the sky in each sub-region, and then perform weighted averaging to calculate the sky illuminance in the region during each period.

[0086] When collecting photovoltaic panel sunlight data, use image denoising algorithms (such as median filtering, bilateral filtering) to remove the noise in the image to ensure the image quality. Convert the first photovoltaic panel image into a photovoltaic panel grayscale image. The grayscale value reflects the brightness of each pixel in the image. Use a semantic segmentation model (such as U-Net, DeepLab, Mask R-CNN, etc.) to segment the image to distinguish the photovoltaic panel from other backgrounds (such as brackets, ground, sky, etc.). Collect photovoltaic panel images with annotations. When annotating, accurately outline the contour of the photovoltaic panel and label each complete photovoltaic panel area. Incomplete photovoltaic panels are ignored. The higher the grayscale value, the greater the light intensity in that area. Calculate the average grayscale value or brightness value of all pixels in the photovoltaic panel area as the average illuminance of the photovoltaic panel. Construct a linear regression model to represent the linear relationship between the grayscale value and the actual illuminance. Establish the mapping relationship between the grayscale value and the illuminance through a calibration experiment. Among them, the linear regression model is expressed as:

[0087] ;

[0088] Among them, and represent adjustment parameters, represents the grayscale value, represents the photovoltaic panel illuminance.

[0089] When collecting the snow data of the photovoltaic panel, collect the latest second photovoltaic panel image data on the second reference day. The camera regularly takes pictures of the photovoltaic panel. When the weather changes (such as snowing, hailing, etc.), increase the shooting frequency to capture the real-time situation. The image preprocessing mainly uses image denoising algorithms (such as median filtering, bilateral filtering) to remove the noise in the image and ensure the image quality. Image enhancement techniques (such as contrast stretching, histogram equalization) can also be used to improve the clarity of the image. Segment the image to identify the boundary of the snow coverage. Perform morphological operations (such as dilation, erosion, opening and closing operations) on the segmentation result to remove small noise areas and smooth the boundary to ensure the segmentation result is more accurate. Use edge detection algorithms (such as Canny edge detection) to further optimize the segmentation result to ensure that the boundary of the snow-covered area is clearly visible. Convert the segmented image into a binary image, where the snow-covered area is white (value is 1) and other areas are black (value is 0). Count the number of white pixels in the binary image. Each pixel represents a unit area. At the same time, obtain the area of the photovoltaic panel. The snow coverage of the photovoltaic panel can be calculated by dividing the number of pixels in the snow-covered area by the number of pixels in the total area of the photovoltaic panel.

[0090] By collecting the latest third photovoltaic panel image data on the second reference day, obtain the dirt data of the photovoltaic panel. Among them, the dirt type is classified by a multi-class classification model (such as ResNet, VGG, EfficientNet, etc.) for the dirt type in the image. The dirt types include: dust, which appears as uniformly distributed fine particles, with a lighter color and stronger reflected light; stain, which appears as local dark spots or stripes, and can be caused by bird droppings, leaves, oil stains, etc.; damage: including cracks, scratches, breakages, etc., which appear as irregular shapes and obvious edges. Set a confidence threshold for each classification result. Only the classification results with high confidence are adopted. The results below the threshold can be marked as "uncertain" and need further manual confirmation. Perform image segmentation identification on the third photovoltaic panel image data. According to the segmentation result, calculate the proportion of the area of the dirty area in the total area of the photovoltaic panel, that is, the weighted sum of the pixel amounts of dust, stain, and damage and the total pixel amount of the photovoltaic panel to obtain the dirtiness of the photovoltaic panel. The weight coefficient of each dirt type can be set according to the actual application scenario or through model training.

[0091] It should be noted that obtaining data such as the sunlight on the photovoltaic panel, the sky sunlight, the snow coverage, and the dirt situation through visual recognition technology can provide richer and more accurate information for photovoltaic power generation prediction. At the same time, if the snow coverage and dirtiness of the photovoltaic panel exceed the preset threshold, the automatic cleaning device can be started to clean the surface of the photovoltaic panel. After the cleaning is completed, collect the photovoltaic panel image data again to obtain the latest snow coverage and dirtiness of the photovoltaic panel. This embodiment can not only improve the power generation efficiency but also optimize the maintenance and management of the system.

[0092] In a preferred embodiment, the calculated influence correction factor data includes:

[0093] Compare the sky illuminance in each time period area on the first reference date with the illuminance corresponding to the total radiation amount in the same time period in the weather forecast on the first reference date. If the average value of the difference between the two is greater than the change threshold, calculate the sky illuminance error through the difference between the two to obtain the first correction factor. If the average value of the difference between the two is less than the change threshold, the first correction factor is set to 1. The change threshold can be set according to the actual application scenario;

[0094] Exemplarily, the sky illuminance error is expressed as:

[0095] ;

[0096] Wherein, represents the number of samples, represents the sky illuminance in each time period area in the weather forecast, represents the sky illuminance in each time period area on the first reference date;

[0097] Then, calculate the first correction factor , which represents the influence of the sky illuminance error on the power generation amount. The formula is as follows:

[0098] ;

[0099] Wherein, represents the maximum allowable sky illuminance error.

[0100] Compare the light intensity of the photovoltaic panel in each time period area on the first reference date with the sky illuminance in each time period area, and calculate the light intensity error of the photovoltaic panel through the difference between the two to obtain the second correction factor;

[0101] Exemplarily, the light intensity error of the photovoltaic panel is expressed as:

[0102] ;

[0103] Wherein, represents the number of samples, represents the sky illuminance in each time period area on the first reference date, represents the light intensity of the photovoltaic panel in each time period area on the first reference date;

[0104] Then, calculate the second correction factor , which represents the influence of the light intensity error of the photovoltaic panel on the power generation amount. The formula is as follows:

[0105] ;

[0106] Among them, represents the maximum allowable light intensity error of the photovoltaic panel.

[0107] Based on the snow coverage degree of the photovoltaic panel, the first influence factor is calculated;

[0108] Exemplarily, the snow coverage degree of the photovoltaic panel obtained by visual recognition technology , with a value range of [0, 1], represents the coverage ratio, and the first influence factor is defined , representing the impact of snow coverage on power generation efficiency. Snow coverage will significantly reduce power generation efficiency. Therefore is a decreasing function and decreases as increases, which is expressed as:

[0109] ;

[0110] Among them, is a constant, representing the degree of influence of snow coverage on power generation efficiency.

[0111] Based on the dirtiness degree of the photovoltaic panel, the second influence factor is calculated;

[0112] Exemplarily, the dirtiness degree of the photovoltaic panel obtained by visual recognition technology , with a value range of [0, 1], represents the dirtiness ratio, and the second influence factor is defined , representing the impact of dirtiness on power generation efficiency. Dirtiness will reduce the light transmittance of the photovoltaic panel, thereby reducing power generation, which is expressed as:

[0113] ;

[0114] Among them, and are constants, representing the degree of influence of dirtiness on power generation efficiency.

[0115] The first correction factor, the second correction factor, the first influence factor, and the second influence factor are combined to generate influence correction factor data.

[0116] In a preferred implementation manner, a composite prediction model is constructed, and the second prediction data for photovoltaic power generation includes:

[0117] Based on the first prediction data and the influence factor data, a composite prediction model is constructed to obtain the corrected second prediction data. The composite prediction model is expressed as:

[0118] ;

[0119] Among them, represents the first prediction data of photovoltaic power generation, represents the first correction factor, represents the second correction factor, represents the first impact factor, represents the second impact factor, represents the dynamic adjustment factor.

[0120] Among them, in order to further improve the prediction accuracy, a dynamic adjustment factor is introduced , and this factor can be adjusted according to factors such as the trend of historical data, seasonal changes, weather patterns, etc. The calculation formula of the dynamic adjustment factor is as follows:

[0121] ;

[0122] wherein, represents the actual power generation in a recent period, represents the predicted power generation in a recent period. The dynamic adjustment factor reflects the deviation of the model and is used to correct future predicted values.

[0123] It should be noted that in this embodiment, through visual recognition and meteorological forecast data, the first prediction data (preliminary prediction based on weather forecast and equipment data) is combined with the impact correction factor data (local environmental information based on visual recognition) to construct a composite prediction model, which can more comprehensively capture various factors affecting photovoltaic power generation. Especially the combination of local information (such as occlusion, dirt) and macro weather conditions can significantly improve the prediction accuracy and is applicable to photovoltaic power stations in complex environments. The introduction of the impact correction factor data enables the model to better cope with extreme weather or emergencies (such as storms, heavy snow, equipment failures, etc.). Even if there is uncertainty in the weather forecast data, the model can still be dynamically adjusted through local environmental information to ensure the stability of the prediction results, effectively reduce the prediction deviation caused by local environmental changes, be closer to the actual power generation, and improve the prediction accuracy.

[0124] The present invention conducts time - series analysis on the historical power generation data of a photovoltaic power station, uses machine learning algorithms (such as LSTM) to predict future power generation trends. Time - series analysis and machine learning techniques are used in the present invention to extract patterns from historical data and make predictions. Through the historical power generation power data and weather data of the photovoltaic power station, a recurrent neural network is used to predict the power generation of the photovoltaic power station. An environmental image around the photovoltaic power station is obtained by using a camera or other image acquisition devices, and these images are analyzed. Image processing and computer vision techniques are used in the present invention to identify factors affecting power generation, such as cloud distribution and light intensity changes. By combining the visual recognition results and historical power data, a composite prediction model is constructed. Data fusion and ensemble learning techniques are used in the present invention to integrate multi - source data, improve the accuracy and robustness of the prediction model, and help the operators of the power system to conduct power dispatching and optimization. Power system dispatching and optimization techniques are used in the present invention to adjust the power generation plan according to the prediction results, and improve the stability and efficiency of the power grid.

[0125] The present invention provides a comprehensive data basis for the model by obtaining multi - source data, enhancing the accuracy of prediction. The model is trained through historical power generation data, and combined with weather forecasts and equipment data for preliminary prediction. Environmental data of the photovoltaic panels are collected through visual recognition technology, key features are extracted and the influence correction factors are calculated to capture the impact of local environmental changes on power generation. By combining the preliminary prediction results with the influence correction factors, a composite prediction model is constructed to obtain a more accurate final power generation prediction, which not only improves the accuracy of prediction, but also enhances the robustness and adaptability of the model, and can better cope with complex and changeable environmental conditions.

[0126] The above is a schematic solution of a composite photovoltaic power generation prediction method based on visual recognition in this embodiment. It should be noted that the technical solution of the composite photovoltaic power generation prediction system based on visual recognition belongs to the same concept as the technical solution of the above - mentioned composite photovoltaic power generation prediction method based on visual recognition. For the details not described in detail in the technical solution of the composite photovoltaic power generation prediction system based on visual recognition in this embodiment, reference can be made to the description of the technical solution of the composite photovoltaic power generation prediction method based on visual recognition.

[0127] The composite photovoltaic power generation prediction system based on visual recognition in this embodiment includes:

[0128] An acquisition module, configured to acquire equipment data, weather forecast data, and historical power generation data of the photovoltaic power station;

[0129] A construction module, configured to construct a photovoltaic power generation prediction model, train the photovoltaic power generation prediction model through historical power generation data, take the weather forecast data and equipment data as inputs, and obtain the first prediction data of photovoltaic power generation;

[0130] A data processing module, configured to collect the environmental data of the photovoltaic panels before the predicted day of the photovoltaic power station, extract key features, and calculate the influence correction factor data;

[0131] An output module, configured to construct a composite prediction model based on the first prediction data and the influence factor data, and obtain the second prediction data of the photovoltaic power generation as the final predicted power of the photovoltaic power generation.

[0132] This embodiment further provides an electronic device applicable to the case of composite photovoltaic power generation prediction based on visual recognition, including:

[0133] A memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for realizing composite photovoltaic power generation prediction based on visual recognition as proposed in the above embodiment.

[0134] This embodiment further provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method for realizing composite photovoltaic power generation prediction based on visual recognition as proposed in the above embodiment.

[0135] The storage medium proposed in this embodiment and the method for realizing composite photovoltaic power generation prediction based on visual recognition proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0136] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (FLASH), a hard disk, or an optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments of the present invention.

[0137] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A composite photovoltaic power generation prediction method based on visual recognition, characterized in that: include: Obtain equipment data, weather forecast data, and historical power generation data of photovoltaic stations; Constructing a photovoltaic power generation prediction model, training the photovoltaic power generation prediction model with the historical power generation data, taking the weather forecast data and equipment data of the prediction day as input, and obtaining first prediction data of photovoltaic power generation; Collect environmental data of photovoltaic panels before the forecast date of the photovoltaic station, extract key features, and calculate the impact correction factor data; Based on the first prediction data and the influencing factor data, a composite prediction model is constructed to obtain second prediction data of photovoltaic power generation as the final photovoltaic power generation prediction power; Environmental data collected before the forecast date for photovoltaic sites include: A plurality of camera points are set around the photovoltaic panels to regularly collect image data around the photovoltaic panels of the photovoltaic station, and environmental data are extracted from the image data through visual recognition, wherein the environmental data includes sky sunshine data, photovoltaic panel sunshine data, photovoltaic panel heavy snow data, and photovoltaic panel dirt data; According to the weather type of the forecast day, a day closest to the forecast day and with the same weather type is determined as the first reference day, and key features of the first reference day are extracted, including photovoltaic panel sunshine light data and sky sunshine light data; The day before the predicted day is taken as the second reference day, and key features of the second reference day are extracted, including photovoltaic panel heavy snow data and photovoltaic panel dirt data.

2. The composite photovoltaic power generation prediction method based on visual recognition according to claim 1, characterized in that: Obtaining weather forecast data and historical power generation data for photovoltaic stations includes: Acquiring equipment data, wherein the equipment data includes photovoltaic panel location, rated power, conversion efficiency, tilt angle, and azimuth; Obtaining weather forecast data for the forecast day, the weather forecast data for the forecast day including time, weather type, total radiation, direct radiation, scattered radiation, reflected radiation, temperature, humidity, and cloud coverage; The historical power generation data for the past one to three years is obtained, wherein the historical power generation data includes actual weather data, equipment data, and actual power generation data of the site.

3. The composite photovoltaic power generation prediction method based on visual recognition according to claim 2, characterized in that: The first prediction data of photovoltaic power generation includes: Preprocessing the equipment data, weather forecast data and historical power generation data, sorting the historical power generation data in time series, and classifying them according to time periodicity based on data date; Dividing the historical power generation data into different training sets, validation sets and test sets according to time period types and weather types; A photovoltaic power generation prediction model is constructed using a long short-term memory network, and an attention mechanism is introduced. The photovoltaic power generation prediction model is trained using the historical power generation data, and the equipment data and weather forecast data are input into the trained photovoltaic power generation prediction model to obtain the first prediction data of the prediction day.

4. The composite photovoltaic power generation prediction method based on visual recognition according to claim 1, characterized in that: The key features extracted include: The sky sunshine light data represents sky illuminance, collects sky images above the photovoltaic panel during the first reference day in each time period, divides the sky images into multiple sub-areas, identifies the sky illuminance of the sub-areas in each time period, and calculates the sky illuminance in the area in each time period; The photovoltaic panel sunshine light data represents the light intensity of the photovoltaic panel, collecting the first photovoltaic panel image data of the first reference day in each time period, converting the first photovoltaic panel image data into a photovoltaic panel grayscale image, segmenting the photovoltaic panel grayscale image, marking each complete photovoltaic panel area, identifying the grayscale value of each photovoltaic panel area, calculating the average grayscale value of the photovoltaic panel, and obtaining the light intensity of the photovoltaic panel in the area in each time period; The photovoltaic panel heavy snow data represents the heavy snow coverage of the photovoltaic panel, collects the latest second photovoltaic panel image data of the second reference day, performs image preprocessing and image segmentation on the second photovoltaic panel image data, converts the segmented image into a binary image, identifies the snow-covered area, counts the number of white pixels in the binary image, and calculates the heavy snow coverage of the photovoltaic panel; The photovoltaic panel contamination data represents the degree of contamination of the photovoltaic panel. The latest third photovoltaic panel image data of the second reference day is collected, and the third photovoltaic panel image data is subjected to image preprocessing and image segmentation. The pixel quantity of different types of contamination in the third photovoltaic panel image data is identified, and the contamination types include dust, stains and damage. The degree of contamination of the photovoltaic panel is obtained based on the weighted sum of the pixel quantity of dust, stains and damage and the total pixel quantity of the photovoltaic panel.

5. The composite photovoltaic power generation prediction method based on visual recognition according to claim 4, characterized in that: The calculated impact correction factor data include: Compare the sky illuminance in each time period of the first reference day with the illuminance corresponding to the total radiation in the same time period in the weather forecast of the first reference day. If the average value of the difference between the two is greater than the change threshold, calculate the sky illuminance error by the difference between the two to obtain a first correction factor. Compare the light intensity of the photovoltaic panel in each time period of the first reference day with the sky light intensity in each time period, and calculate the light intensity error of the photovoltaic panel by the difference between the two to obtain a second correction factor; Based on the snow coverage of the photovoltaic panel, a first influencing factor is calculated; Based on the degree of contamination of the photovoltaic panel, a second influencing factor is calculated; The first correction factor, the second correction factor, the first impact factor and the second impact factor are combined to generate impact correction factor data.

6. The composite photovoltaic power generation prediction method based on visual recognition according to claim 5, characterized in that: The second prediction data of photovoltaic power generation includes: Based on the first prediction data and the influencing factor data, a composite prediction model is constructed to obtain the revised second prediction data. The composite prediction model is expressed as: ； in, represents the first forecast data of photovoltaic power generation, represents the first correction factor, represents the second correction factor, represents the first impact factor, represents the second impact factor, Represents the dynamic adjustment factor.

7. A composite photovoltaic power generation prediction system based on visual recognition, used in the composite photovoltaic power generation prediction method based on visual recognition as claimed in any one of claims 1 to 6, characterized in that: include, The acquisition module is used to obtain equipment data, weather forecast data and historical power generation data of photovoltaic stations; A construction module is used to construct a photovoltaic power generation prediction model, train the photovoltaic power generation prediction model through the historical power generation data, take the weather forecast data and the equipment data as input, and obtain the first prediction data of photovoltaic power generation; The data processing module is used to collect the environmental data of photovoltaic panels before the forecast date of the photovoltaic station, extract key features, and calculate the impact correction factor data; The output module is used to construct a composite prediction model based on the first prediction data and the influencing factor data to obtain the second prediction data of photovoltaic power generation as the final photovoltaic power generation prediction power.

8. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the composite photovoltaic power generation prediction method based on visual recognition according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, implement the steps of the composite photovoltaic power generation prediction method based on visual recognition as described in any one of claims 1 to 6.

Images