Automatic photovoltaic panel cleaning method and system combining weather and operation data analysis

Abstract

The invention discloses a photovoltaic panel automatic cleaning method and system combined with weather and operation data analysis. The photovoltaic panel automatic cleaning method comprises the steps that first historical data and second historical data of a target photovoltaic panel are obtained; establishing a first efficiency model and a second rainfall influence model according to the first historical data and the second historical data; establishing a first judgment strategy based on the first efficiency model and the second rainfall influence model; and first real-time data and second real-time data of the target photovoltaic panel are obtained, and the target photovoltaic panel is automatically cleaned in combination with the first judgment strategy. By monitoring and analyzing the weather conditions and the operation data of the photovoltaic panel in real time, the opportunity and the mode of cleaning can be intelligently determined, so that the overall efficiency of the photovoltaic system is improved. Unnecessary cleaning operation can be reduced, maintenance cost is saved, meanwhile, it is ensured that the photovoltaic panel operates in the optimal state, and the service life of the photovoltaic panel is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic cleaning of photovoltaic panels, and in particular to an automatic cleaning method and system for photovoltaic panels combined with weather and operation data analysis. Background Art

[0002] In the current field of photovoltaic panel cleaning technology, the commonly used methods are timed cleaning or cleaning after manual observation. However, these traditional methods have many disadvantages, such as low efficiency, high cost and inability to respond to environmental changes in real time.

[0003] Specifically, the scheduled cleaning method may cause unnecessary consumption of resources when the photovoltaic panels do not actually need to be cleaned, and when cleaning is really needed to maintain optimal power generation efficiency, the best cleaning time may be missed due to the limitations of the scheduled plan.

[0004] In addition, manual observation is limited by human subjective judgment and work intensity, which makes it difficult to ensure the timeliness and accuracy of cleaning work. Since manual observation relies on the subjective judgment of staff, misjudgment may occur, resulting in untimely or excessive cleaning, thus wasting resources. At the same time, manual cleaning work is intensive, especially in large-scale photovoltaic power stations, which require a large amount of manpower for cleaning work, which not only increases labor costs, but also may not guarantee the timeliness of cleaning work due to limited human resources.

[0005] Therefore, existing cleaning methods have many shortcomings in practical applications, and there is an urgent need for a more efficient, intelligent cleaning technology that can respond to environmental changes in real time to solve these problems. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the present invention provides a method and system for automatically cleaning photovoltaic panels in combination with weather and operation data analysis, which can solve the problems mentioned in the background technology.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] In a first aspect, the present invention provides a method for automatically cleaning a photovoltaic panel by combining weather and operation data analysis, comprising:

[0011] Acquire first historical data and second historical data of a target photovoltaic panel;

[0012] Establishing a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data;

[0013] Establishing a first judgment strategy based on the first efficiency model and the second rainfall impact model;

[0014] The first real-time data and the second real-time data of the target photovoltaic panel are acquired, and the target photovoltaic panel is automatically cleaned in combination with the first judgment strategy.

[0015] As a preferred solution of the photovoltaic panel automatic cleaning method combined with weather and operation data analysis described in the present invention, wherein: the establishment of the first efficiency model and the second rainfall impact model according to the first historical data and the second historical data includes:

[0016] The second historical data at least includes historical data of sunshine intensity and historical data of rainfall of the target photovoltaic panel;

[0017] The first efficiency model is an arbitrary model whose input is the first historical data and the historical data of sunshine intensity, and whose output is the photovoltaic operation efficiency;

[0018] The second rainfall impact model is an arbitrary model whose input is the first historical data and the second historical data, and whose output is the impact of the rainfall data on the photovoltaic operation efficiency.

[0019] As a preferred solution of the photovoltaic panel automatic cleaning method combining weather and operation data analysis described in the present invention, the first judgment strategy includes:

[0020] Acquire a first efficiency threshold and a second efficiency threshold according to the first efficiency model;

[0021] A first rainfall impact threshold and a second rainfall impact threshold are acquired according to the second rainfall impact model.

[0022] As a preferred solution of the photovoltaic panel automatic cleaning method combined with weather and operation data analysis described in the present invention, wherein: the acquisition of the first real-time data and the second real-time data of the target photovoltaic panel, and the automatic cleaning of the target photovoltaic panel in combination with the first judgment strategy includes:

[0023] The first real-time data corresponds to the first historical data;

[0024] The second real-time data corresponds to the second historical data;

[0025] The second real-time data includes at least real-time data on the sunshine intensity and real-time data on the rainfall of the target photovoltaic panel.

[0026] As a preferred solution of the photovoltaic panel automatic cleaning method combined with weather and operation data analysis described in the present invention, wherein: the acquisition of the first real-time data and the second real-time data of the target photovoltaic panel, and the automatic cleaning of the target photovoltaic panel in combination with the first judgment strategy also includes:

[0027] Acquire the real-time rainfall data in the second real-time data, and compare the real-time rainfall data with the first rainfall impact threshold;

[0028] If the real-time rainfall data is less than the first rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity;

[0029] If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, primary cleaning is performed;

[0030] If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold;

[0031] If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, perform level 3 cleaning;

[0032] If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

[0033] As a preferred solution of the photovoltaic panel automatic cleaning method combined with weather and operation data analysis described in the present invention, wherein: the acquisition of the first real-time data and the second real-time data of the target photovoltaic panel, and the automatic cleaning of the target photovoltaic panel in combination with the first judgment strategy also includes:

[0034] If the real-time rainfall data is not less than the first rainfall impact threshold and less than the second rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity;

[0035] If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, secondary cleaning is performed;

[0036] If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold;

[0037] If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, a fourth-level cleaning is performed;

[0038] If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

[0039] As a preferred scheme of the method for automatic cleaning of photovoltaic panels combined with weather and operation data analysis described in the present invention, wherein: the obtaining of the first real-time data and the second real-time data of the target photovoltaic panel, and the automatic cleaning of the target photovoltaic panel in combination with the first judgment strategy also includes: if the real-time rainfall data is not less than the second rainfall impact threshold, no cleaning operation is performed.

[0040] In a second aspect, the present invention provides a photovoltaic panel automatic cleaning system combining weather and operation data analysis, comprising:

[0041] A data acquisition module, used to acquire first historical data and second historical data of a target photovoltaic panel;

[0042] A model building module, used to build a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data;

[0043] A strategy generation module, used to establish a first judgment strategy based on the first efficiency model and the second rainfall impact model;

[0044] The judgment module is used to obtain the first real-time data and the second real-time data of the target photovoltaic panel, and automatically clean the target photovoltaic panel in combination with the first judgment strategy.

[0045] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method when executing the computer program.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method described above when executed by a processor.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention proposes a photovoltaic panel automatic cleaning method and system combining weather and operation data analysis, obtaining the first historical data and the second historical data of the target photovoltaic panel; establishing the first efficiency model and the second rainfall impact model according to the first historical data and the second historical data; establishing the first judgment strategy based on the first efficiency model and the second rainfall impact model; obtaining the first real-time data and the second real-time data of the target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy. By real-time monitoring and analysis of weather conditions and the operation data of the photovoltaic panel, the timing and method of cleaning can be intelligently determined, thereby improving the overall efficiency of the photovoltaic system. It can reduce unnecessary cleaning operations, save maintenance costs, and ensure that the photovoltaic panel operates in the best state, extending its service life. Since the cleaning operation is based on accurate data models and threshold judgments, potential damage to the photovoltaic panel caused by excessive cleaning or insufficient cleaning can be avoided. The implementation of the present application not only improves the energy output of the photovoltaic system, but also reduces the operating cost of the photovoltaic power station by reducing the number of cleaning times and optimizing the maintenance process. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0049] Figure 1 A method flow chart of a photovoltaic panel automatic cleaning method and system combining weather and operation data analysis provided by one embodiment of the present invention;

[0050] Figure 2 An internal structural diagram of a computer device of a photovoltaic panel automatic cleaning method and system combining weather and operation data analysis provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0052] Example 1

[0053] Reference Figure 1-Figure 2, which is the first embodiment of the present invention, provides a photovoltaic panel automatic cleaning method and system combining weather and operation data analysis, including:

[0054] There are some problems in the existing related technologies, such as inappropriate selection of cleaning timing leading to low efficiency of photovoltaic panels, and too high cleaning frequency causing waste of resources.

[0055] The present application provides a method that can effectively solve the above-mentioned problems. Next, multiple embodiments will be combined to explain in detail how to implement the photovoltaic panel automatic cleaning method combined with weather and operation data analysis;

[0056] Figure 1 A method flow chart of a photovoltaic panel automatic cleaning method and system combining weather and operation data analysis is shown, including:

[0057] S101, acquiring first historical data and second historical data of a target photovoltaic panel;

[0058] In an optional embodiment, the first historical data of the target photovoltaic panel may include photovoltaic operating parameters, such as the power generation, operating temperature, fault records, operating status (such as whether it is in working mode), energy conversion efficiency, efficiency comparison before and after cleaning, etc. The second historical data of the target photovoltaic panel may include weather type data of the area where the target photovoltaic panel is located, such as historical records of weather conditions such as sunny, cloudy, overcast, and rainy days, and may also include historical data of meteorological parameters such as temperature, humidity, wind speed, and wind direction, and may also include specific data such as sunshine intensity and sunshine duration.

[0059] In an optional embodiment, the purpose of obtaining the first historical data is to accurately obtain the operating efficiency of the target photovoltaic panel, so any first historical data that can be combined with the sunshine intensity historical data to obtain the operating efficiency of the target photovoltaic panel can be selected;

[0060] In an optional embodiment, the efficiency of the photovoltaic panel under different lighting conditions can be evaluated by analyzing the power generation of the photovoltaic panel under different sunlight intensities.

[0061] In an optional embodiment, the relationship between the operating temperature and the power generation efficiency can be analyzed to understand the impact of high or low temperature on the performance of the photovoltaic panel.

[0062] In an embodiment of the present application, the first historical data of the target photovoltaic panel selects the power generation and operating temperature historical data of the photovoltaic panel, and the second historical data selects the sunshine intensity historical data for the power generation and operating temperature historical data of the photovoltaic panel.

[0063] It should be noted that obtaining the first historical data and the second historical data of the target photovoltaic panel can provide the necessary data support for establishing the efficiency model. By analyzing the historical data, the operating efficiency of the photovoltaic panel under different weather conditions can be determined, thereby providing a scientific basis for the subsequent cleaning strategy. In addition, obtaining the second historical data of the target photovoltaic panel, especially the data on the weather type, is helpful to establish a rainfall impact model, which can predict the impact of rainfall on the power generation efficiency of the photovoltaic panel and provide decision support for the automatic cleaning system. Through these models, the system can intelligently determine when to clean, as well as the frequency and method of cleaning, to ensure that the photovoltaic panel operates in the best condition, while avoiding efficiency loss and additional maintenance costs caused by improper cleaning.

[0064] S102, establishing a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data;

[0065] In the embodiment of the present application, establishing the first efficiency model and the second rainfall impact model according to the first historical data and the second historical data includes:

[0066] The second historical data includes at least the sunshine intensity historical data and the rainfall historical data of the target photovoltaic panel;

[0067] The first efficiency model is an arbitrary model whose input is the first historical data and the sunshine intensity historical data, and whose output is the photovoltaic operation efficiency;

[0068] The second rainfall impact model is an arbitrary model whose input is the first historical data and the second historical data, and whose output is the impact of the rainfall data on the photovoltaic operation efficiency.

[0069] In an optional embodiment, the first efficiency model can be established by a fitting method, by collecting the first historical data (such as power generation, operating temperature, etc.) of the target photovoltaic panel and the historical data of sunshine intensity, and using a mathematical fitting method (such as linear regression, polynomial fitting, etc.) to establish a model. The fitting method can help us find the functional relationship between the input variables (such as sunshine intensity, temperature, etc.) and the output variables (such as photovoltaic operating efficiency). For example, the least squares method can be used to fit the historical data to obtain a mathematical model that can predict the operating efficiency of the photovoltaic panel.

[0070] In an optional embodiment, the first efficiency model can also classify similar data points into one category by clustering historical data, thereby discovering the operating efficiency characteristics of photovoltaic panels under different conditions. Cluster analysis can help identify which factors have a greater impact on photovoltaic efficiency. For example, the K-means clustering algorithm can be used to classify data under different sunshine intensity and temperature conditions, and then analyze the average efficiency of each type of data.

[0071] In an optional embodiment, the first efficiency model can also use a neural network (such as a multi-layer perceptron, a convolutional neural network, etc.) to train historical data to establish a model that can predict the operating efficiency of photovoltaic panels. Neural networks have powerful nonlinear modeling capabilities and can capture complex input-output relationships. For example, a deep learning framework (such as TensorFlow, PyTorch, etc.) can be used to build a neural network model with the first historical data and sunshine intensity historical data as input and the photovoltaic operating efficiency as output.

[0072] In the embodiment of the present application, the first efficiency model is established by a fitting method, and the fitting method is a new improved fitting method, which is as follows:

[0073] Collect the first historical data of the target photovoltaic panel (such as power generation, operating temperature, fault records, etc., this application selects power generation and operating temperature) and the historical data of sunshine intensity.

[0074] Furthermore, clean the data to remove outliers and missing values. Standardize the data to ensure that all features are on the same scale. Select features that have a greater impact on photovoltaic efficiency.

[0075] Furthermore, a polynomial fitting method is used to build a preliminary model, and a regularization term (such as L2 regularization) is introduced to prevent overfitting. The K-fold cross-validation method is used to train and verify the model, and the grid search method is used to find the optimal model parameters.

[0076] Furthermore, the mean square error (MSE) and coefficient of determination (R2) of the model were calculated, the residual distribution of the model was analyzed, and whether the model had systematic bias was checked.

[0077] Furthermore, according to the model's prediction results and actual operation data, the model parameters are dynamically adjusted, and a variety of fitting methods are combined and an integrated learning method is used to further improve the model's prediction accuracy and stability.

[0078] For example, assuming that the original data is (P i ,I i ,T i ), where P i is the power generation, I i is the sunlight intensity T i It's the temperature.

[0079] For example, we can use the Z-score method to detect and remove outliers:

[0080]

[0081] Among them, μ P and σ Pare the mean and standard deviation of power generation respectively. Usually, when |z i |>3, it is considered that P i These are outliers and need to be removed.

[0082] Furthermore, the data is normalized to have the same scale. A common method is Z-score normalization:

[0083]

[0084] Among them, μ P ,μ I ,μ T and σ P ,σ I ,σ T are the mean and standard deviation of power generation, sunshine intensity and temperature respectively. Furthermore, it is assumed that there is a polynomial relationship between power generation P, sunshine intensity I and temperature T:

[0085] P=β0+β1I+β2T+β3I 2 +β4IT+β5T 2 +∈

[0086] Among them, β0, β1, β2, β3, β4, β5 are the coefficients to be estimated, and ∈ is the error term.

[0087] Furthermore, based on polynomial fitting, the L2 regularization term is introduced to prevent overfitting.

[0088]

[0089] Among them, In is the regularization parameter, which needs to be determined by methods such as cross-validation.

[0090] Furthermore, the K-fold cross validation method is used to train and validate the model. The data set is divided into K subsets, and K-1 subsets are used for training each time, and the remaining 1 subset is used for validation. This is repeated K times, and the average performance index is taken.

[0091] Furthermore, the grid search method is used to find the optimal model parameters. For example, different polynomial orders and regularization parameters λ can be set, and the best combination can be selected through cross-validation.

[0092] Furthermore, we calculate the model prediction value The mean square error between the actual value P is:

[0093]

[0094] Furthermore, the coefficient of determination R of the model was calculated. 2 , evaluate the model's ability to explain the data:

[0095]

[0096] in, is the mean value of actual power generation.

[0097] Furthermore, the residual distribution of the model is analyzed to check whether the model has systematic deviations. The residual is defined as:

[0098]

[0099] Furthermore, the model parameters can be adjusted dynamically based on the model's prediction results and actual operation data to continuously optimize the model performance. For example, online learning methods can be used to gradually update the model parameters.

[0100] In an optional embodiment, the second rainfall impact model can be established by a fitting method, similar to the establishment of the first efficiency model, and a mathematical fitting method (such as linear regression, polynomial fitting, etc.) can be used to establish the second rainfall impact model. By collecting the first historical data (such as power generation, operating temperature, etc.) and the second historical data (such as rainfall historical data) of the target photovoltaic panel, a mathematical fitting method is used to establish a model to help find the functional relationship between the input variables (such as rainfall, sunshine intensity, etc.) and the output variables (such as changes in photovoltaic operation efficiency).

[0101] In an optional embodiment, the second rainfall impact model can also cluster historical data according to rainfall size, classify similar data points into one category, and thus discover the characteristics of photovoltaic panel operating efficiency under different rainfall conditions. For example, the K-means clustering algorithm is used to classify data under different rainfall conditions, and then the average efficiency change of each type of data is analyzed to evaluate the impact of different rainfall on photovoltaic efficiency.

[0102] In an optional embodiment, the second rainfall impact model can also use time series analysis methods (such as ARIMA model, LSTM, etc.) to predict the impact of future rainfall on the operating efficiency of photovoltaic panels. This method is particularly suitable for processing time-dependent data and can better capture the long-term impact of rainfall changes on photovoltaic efficiency.

[0103] In an embodiment of the present application, the second rainfall impact model trains historical data using a neural network (such as a multi-layer perceptron, a convolutional neural network, etc.) to establish a model that can predict the impact of rainfall on photovoltaic operation efficiency. Neural networks have powerful nonlinear modeling capabilities and can capture complex input-output relationships. For example, a deep learning framework (such as TensorFlow, PyTorch, etc.) can be used to build a neural network model, with the input being the first historical data and the second historical data (such as rainfall, sunshine intensity, etc.), and the output being the impact of rainfall on photovoltaic operation efficiency.

[0104] Exemplarily, historical data is collected, and the first historical data may include the power generation, operating temperature, fault records, etc. of the photovoltaic panel. The second historical data may include weather data of the area where the target photovoltaic panel is located, such as rainfall, sunshine intensity, temperature, humidity, wind speed, wind direction, etc.

[0105] Furthermore, outliers and missing values ​​can be removed. The Z-score method can be used to detect and remove outliers.

[0106]

[0107] Among them, X is a certain feature value, μ and σ are the mean and standard deviation of the feature respectively. Usually, when |Z|>3, the value is considered to be an outlier and needs to be removed.

[0108] Furthermore, the data is normalized to have the same scale. A common method is Z-score normalization:

[0109]

[0110] Among them, X is a certain eigenvalue, μ and σ are the mean and standard deviation of the feature respectively.

[0111] Furthermore, characteristics that have a greater impact on photovoltaic efficiency are selected. This application selects rainfall, sunshine intensity, temperature, etc.

[0112] Going further, choose the neural network architecture: you can choose a multi-layer perceptron (MLP), a convolutional neural network (CNN), etc. For time series data, you can consider using a long short-term memory network (LSTM).

[0113] Furthermore, the input is the first historical data (such as power generation, operating temperature, etc.) and the second historical data (such as rainfall, sunshine intensity, etc.), and the output is the impact of rainfall on photovoltaic operation efficiency.

[0114] Going a step further, divide the dataset into training set, validation set and test set. The common division ratio is 70% training set, 15% validation set, and 15% test set.

[0115] Furthermore, we select the loss function and optimizer. Loss function: Mean Squared Error (MSE):

[0116]

[0117] Furthermore, the optimizer selects Adam, RMSprop, etc.

[0118] Furthermore, the model is trained using the training set, validated using the validation set, and hyperparameters (such as learning rate, number of hidden layer nodes, batch size, etc.) are adjusted to optimize model performance.

[0119] Furthermore, the K-fold cross validation method is used to train and validate the model. The data set is divided into K subsets, and K-1 subsets are used for training each time, and the remaining 1 subset is used for validation. This is repeated K times, and the average performance index is taken.

[0120] Furthermore, we calculate the performance metric, mean square error (MSE):

[0121]

[0122] Coefficient of determination R 2 :

[0123]

[0124] in, is the mean of the actual values. 2 The closer the value is to 1, the better the model fits the data, that is, the higher the proportion of variation explained by the model to the total variation. 2 The closer the value is to 0, the worse the model fits the data.

[0125] Furthermore, we check whether the model has systematic bias. The residual is defined as:

[0126]

[0127] Furthermore, the model parameters can be adjusted dynamically based on the model's prediction results and actual operation data to continuously optimize the model performance. For example, online learning methods can be used to gradually update the model parameters;

[0128] Furthermore, the trained model is deployed in practical applications, and the impact of rainfall on photovoltaic operation efficiency is predicted in combination with real-time data to provide decision support for the automatic cleaning system.

[0129] It should be noted that establishing the first efficiency model and the second rainfall impact model based on the first historical data and the second historical data can more accurately predict the operating efficiency of the photovoltaic panel under different weather conditions. In this way, the automatic cleaning system can more intelligently determine the timing and frequency of cleaning, thereby reducing unnecessary cleaning times while ensuring power generation efficiency and extending the service life of the photovoltaic panel. In addition, the method can also adjust the cleaning plan in advance based on real-time weather forecast data to cope with upcoming weather changes and ensure that the photovoltaic panel is always in the best working condition.

[0130] S103, establishing a first judgment strategy based on the first efficiency model and the second rainfall impact model;

[0131] In the embodiment of the present application, the first judgment strategy includes:

[0132] Acquire a first efficiency threshold and a second efficiency threshold according to a first efficiency model;

[0133] A first rainfall impact threshold and a second rainfall impact threshold are obtained according to the second rainfall impact model.

[0134] In an optional embodiment, relevant technicians can increase or decrease the number of corresponding thresholds according to actual needs. In the present application, two efficiency thresholds and two rainfall thresholds are designed due to different methods of establishing the model.

[0135] In an optional embodiment, after the first efficiency model is established, the first efficiency threshold and the second efficiency threshold can be determined by analyzing the output results of the model. These two thresholds are used to determine whether the current operating efficiency of the photovoltaic panel is within a normal range. For example, the first efficiency threshold can be set to a lower limit of the normal operating efficiency, and the second efficiency threshold can be set to a higher limit.

[0136] In an optional embodiment, in the process of establishing the second rainfall impact model, the changes in the operating efficiency of the photovoltaic panels under different rainfall conditions in the historical data are analyzed. Through statistical analysis or fitting of these data, the key points of the impact of rainfall on the efficiency of the photovoltaic panels can be determined, and then the first rainfall impact threshold and the second rainfall impact threshold can be set. For example, the first rainfall impact threshold can be set to a rainfall value that has a smaller impact on the efficiency of the photovoltaic panels, and the second rainfall impact threshold can be set to a rainfall value that has a greater impact on the efficiency of the photovoltaic panels.

[0137] In the present application example, the first efficiency threshold is set to a lower limit of the normal operating efficiency. For example, by analyzing the model output, assuming that the average value of the normal operating efficiency is 0.85 and the standard deviation is 0.05, the first efficiency threshold can be set to 0.80 (i.e., the average value minus 1 standard deviation). The second efficiency threshold: is set to a higher limit of the normal operating efficiency. For example, the second efficiency threshold can be set to 0.90 (i.e., the average value plus 1 standard deviation).

[0138] In the present application example, the first rainfall impact threshold is set to a rainfall value that has a smaller impact on the efficiency of the photovoltaic panel. For example, by analyzing the model output, assuming that the rainfall has a smaller impact on the efficiency of the photovoltaic panel when it is 0-5 mm, the first rainfall impact threshold can be set to 5 mm. The second rainfall impact threshold is set to a rainfall value that has a greater impact on the efficiency of the photovoltaic panel. For example, assuming that the rainfall has a greater impact on the efficiency of the photovoltaic panel when it is above 10 mm, the second rainfall impact threshold can be set to 10 mm.

[0139] It should be noted that the specific threshold is not limited in this application. After the relevant technical personnel carry out the operation modeling of this application, the output results obtained can be intuitively seen or the specific threshold size can be set. There will also be different threshold values ​​for different photovoltaic panels.

[0140] It should be noted that establishing the first judgment strategy based on the first efficiency model and the second rainfall impact model can provide a more accurate decision support.

[0141] S104, acquiring first real-time data and second real-time data of a target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with a first judgment strategy.

[0142] In the example of the present application, obtaining the first real-time data and the second real-time data of the target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy includes:

[0143] The first real-time data corresponds to the first historical data;

[0144] The second real-time data corresponds to the second historical data;

[0145] The second real-time data includes at least the real-time data of sunshine intensity and rainfall of the target photovoltaic panel.

[0146] In the example of the present application, obtaining the first real-time data and the second real-time data of the target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy also includes:

[0147] Acquire the real-time rainfall data in the second real-time data, and compare the real-time rainfall data with the first rainfall impact threshold;

[0148] If the real-time rainfall data is less than the first rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity;

[0149] If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, primary cleaning is performed;

[0150] If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, then determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold;

[0151] If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, the third-level cleaning is performed;

[0152] If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

[0153] In the example of the present application, obtaining the first real-time data and the second real-time data of the target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy also includes:

[0154] If the real-time rainfall data is not less than the first rainfall impact threshold and less than the second rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity;

[0155] If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, secondary cleaning is performed;

[0156] If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, then determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold;

[0157] If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, the fourth level cleaning is performed;

[0158] If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

[0159] In the example of the present application, the first real-time data and the second real-time data of the target photovoltaic panel are obtained, and the automatic cleaning of the target photovoltaic panel in combination with the first judgment strategy also includes: if the real-time rainfall data is not less than the second rainfall impact threshold, no cleaning operation is performed.

[0160] In an optional embodiment, a first efficiency change threshold and a second efficiency change threshold (the first efficiency change threshold is greater than the second efficiency change threshold) can also be obtained based on experimental operations. After the photovoltaic panel undergoes secondary cleaning, tertiary cleaning, and quaternary cleaning, the relationship between the difference between the efficiency before cleaning and the efficiency after cleaning and the first efficiency change threshold and the second efficiency change threshold can be determined.

[0161] Example 2

[0162] This embodiment also provides some optional embodiments, including:

[0163] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the fourth-level cleaning is greater than the first efficiency change value, it means that the photovoltaic panel has not been cleaned for a long time, so a large frequency increase will occur after the fourth-level cleaning. At this time, wait for the fourth-level cleaning to be completed and then perform the first-level cleaning again;

[0164] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the fourth-level cleaning is not greater than the first efficiency change value, and greater than the second efficiency change value, it means that the photovoltaic panel has not been cleaned for a period of time, so a relatively large frequency increase will occur after the fourth-level cleaning. At this time, wait for the fourth-level cleaning to be completed and then re-perform the second-level cleaning;

[0165] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the fourth-level cleaning is not greater than the second efficiency change value, it means that there is no significant efficiency change after the cleaning operation, so repeated cleaning is not performed.

[0166] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the third-level cleaning is greater than the first efficiency change value, it means that the photovoltaic panel has not been cleaned for a long time, so a large frequency increase will occur only after the third-level cleaning. At this time, wait for the third-level cleaning to be completed and then perform the first-level cleaning again;

[0167] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the third-level cleaning is not greater than the first efficiency change value, and greater than the second efficiency change value, it means that the photovoltaic panel has not been cleaned for a period of time, so a relatively large frequency increase will occur after the third-level cleaning. At this time, wait for the third-level cleaning to be completed and then re-perform the second-level cleaning;

[0168] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the third-level cleaning is not greater than the second efficiency change value, it means that there is no significant efficiency change after the cleaning operation, so repeated cleaning is not performed.

[0169] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the secondary cleaning is greater than the first efficiency change value, it means that the photovoltaic panel has not been cleaned for a long time, so a large frequency increase will occur after the secondary cleaning. At this time, wait for the secondary cleaning to be completed and then perform the primary cleaning again;

[0170] In an optional embodiment, when the value of the increase in the efficiency of the photovoltaic panel after the secondary cleaning is not greater than the first efficiency change value, repeated cleaning is not performed.

[0171] In an optional embodiment, level 1 cleaning may be the most thorough cleaning operation, suitable for situations where the photovoltaic panel has not been cleaned for a long time and the efficiency has dropped significantly. Level 1 cleaning may include the use of professional cleaning equipment and chemical solvents, and may even require the disassembly of some components for deep cleaning. Level 2 cleaning may be performed when the efficiency of the photovoltaic panel has dropped significantly, but not to the extent that level 1 cleaning is required. This may include the use of high-pressure water guns and professional cleaning agents to thoroughly remove stubborn dirt and deposits. Level 3 cleaning may be performed when the efficiency of the photovoltaic panel has dropped to a certain extent, but it is still not enough to trigger a higher level of cleaning. This usually involves the use of water and mild detergents to remove accumulated dirt, but will not use tools such as high-pressure water guns that may cause physical damage. Level 4 cleaning may be the mildest cleaning operation, usually performed when the efficiency of the photovoltaic panel has dropped slightly, but not to a serious level. Level 4 cleaning may only involve the use of a soft brush or airflow to remove surface dust.

[0172] In an optional embodiment, the implementation of primary cleaning, secondary cleaning, tertiary cleaning and quaternary cleaning is based on real-time monitoring of photovoltaic panel efficiency and analysis of historical data. By setting different efficiency thresholds and efficiency change thresholds, the system can intelligently determine the level and frequency of cleaning to ensure that the cleaning efficiency and energy output of the photovoltaic panel are maximized. In addition, the system can also automatically adjust the cleaning strategy based on the real-time and historical data of the photovoltaic panel to adapt to different weather conditions and operating conditions. For example, in rainy seasons, the system may reduce the frequency of cleaning, while in dry seasons, it may increase the number of cleanings to maintain the high efficiency of the photovoltaic panel. Through this intelligent management, the service life of the photovoltaic panel can be effectively extended while reducing maintenance costs.

[0173] In summary, the present invention proposes a method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis, obtaining first historical data and second historical data of a target photovoltaic panel; establishing a first efficiency model and a second rainfall impact model based on the first historical data and the second historical data; establishing a first judgment strategy based on the first efficiency model and the second rainfall impact model; obtaining first real-time data and second real-time data of a target photovoltaic panel, and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy. By real-time monitoring and analysis of weather conditions and operation data of photovoltaic panels, the timing and method of cleaning can be intelligently determined, thereby improving the overall efficiency of the photovoltaic system. It can reduce unnecessary cleaning operations, save maintenance costs, and ensure that photovoltaic panels operate in the best state, extending their service life. Since the cleaning operation is based on accurate data models and threshold judgments, potential damage to photovoltaic panels caused by excessive or insufficient cleaning can be avoided. The implementation of this application not only improves the energy output of the photovoltaic system, but also reduces the operating costs of photovoltaic power stations by reducing the number of cleaning times and optimizing maintenance processes.

[0174] Example 3

[0175] This embodiment also provides a photovoltaic panel automatic cleaning system combining weather and operation data analysis, including:

[0176] A data acquisition module, used to acquire first historical data and second historical data of a target photovoltaic panel;

[0177] A model building module, used to build a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data;

[0178] A strategy generation module, used for establishing a first judgment strategy based on a first efficiency model and a second rainfall impact model;

[0179] The judgment module is used to obtain the first real-time data and the second real-time data of the target photovoltaic panel, and automatically clean the target photovoltaic panel in combination with the first judgment strategy.

[0180] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0181] This embodiment also provides a computer device, which may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 2As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for automatically cleaning a photovoltaic panel combined with weather and operation data analysis is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0182] This embodiment further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0183] Acquire first historical data and second historical data of a target photovoltaic panel;

[0184] Establishing a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data;

[0185] Establishing a first judgment strategy based on a first efficiency model and a second rainfall impact model;

[0186] The first real-time data and the second real-time data of the target photovoltaic panel are acquired, and the target photovoltaic panel is automatically cleaned in combination with the first judgment strategy.

[0187] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0188] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0189] 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.

[0190] 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 A function specified in one or more boxes.

[0191] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0192] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0193] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for automatically cleaning photovoltaic panels by combining weather and operation data analysis, characterized in that: include: Acquire first historical data and second historical data of a target photovoltaic panel; Establishing a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data; Establishing a first judgment strategy based on the first efficiency model and the second rainfall impact model; The first real-time data and the second real-time data of the target photovoltaic panel are acquired, and the target photovoltaic panel is automatically cleaned in combination with the first judgment strategy.

2. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 1, characterized in that: The establishing of the first efficiency model and the second rainfall impact model according to the first historical data and the second historical data comprises: The second historical data at least includes historical data of sunshine intensity and historical data of rainfall of the target photovoltaic panel; The first efficiency model is an arbitrary model whose input is the first historical data and the historical data of sunshine intensity, and whose output is the photovoltaic operation efficiency; The second rainfall impact model is an arbitrary model whose input is the first historical data and the second historical data, and whose output is the impact of the rainfall data on the photovoltaic operation efficiency.

3. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 2, characterized in that: The first judgment strategy includes: Acquire a first efficiency threshold and a second efficiency threshold according to the first efficiency model; A first rainfall impact threshold and a second rainfall impact threshold are acquired according to the second rainfall impact model.

4. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 3, characterized in that: The step of acquiring the first real-time data and the second real-time data of the target photovoltaic panel and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy includes: The first real-time data corresponds to the first historical data; The second real-time data corresponds to the second historical data; The second real-time data includes at least real-time data on the sunshine intensity and real-time data on the rainfall of the target photovoltaic panel.

5. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 4, characterized in that: The acquiring of the first real-time data and the second real-time data of the target photovoltaic panel and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy further comprises: Acquire the real-time rainfall data in the second real-time data, and compare the real-time rainfall data with the first rainfall impact threshold; If the real-time rainfall data is less than the first rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity; If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, primary cleaning is performed; If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold; If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, perform level 3 cleaning; If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

6. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 5, characterized in that: The acquiring of the first real-time data and the second real-time data of the target photovoltaic panel and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy further comprises: If the real-time rainfall data is not less than the first rainfall impact threshold and less than the second rainfall impact threshold, the relationship between the real-time efficiency of the photovoltaic panel and the first efficiency threshold is determined based on the first real-time data and the real-time data of sunshine intensity; If the real-time efficiency of the photovoltaic panel is less than the first efficiency threshold, secondary cleaning is performed; If the real-time efficiency of the photovoltaic panel is not less than the first efficiency threshold, determining the magnitude relationship between the real-time efficiency of the photovoltaic panel and the second efficiency threshold; If the real-time efficiency of the photovoltaic panel is less than the second efficiency threshold, a fourth-level cleaning is performed; If the real-time efficiency of the photovoltaic panel is not less than the second efficiency threshold, no cleaning operation is performed.

7. The method for automatically cleaning photovoltaic panels in combination with weather and operation data analysis as claimed in claim 6, characterized in that: The acquiring of the first real-time data and the second real-time data of the target photovoltaic panel and automatically cleaning the target photovoltaic panel in combination with the first judgment strategy also includes: if the real-time rainfall data is not less than the second rainfall impact threshold, no cleaning operation is performed.

8. An automatic photovoltaic panel cleaning system combining weather and operation data analysis, characterized in that: include: A data acquisition module, used to acquire first historical data and second historical data of a target photovoltaic panel; A model building module, used to build a first efficiency model and a second rainfall impact model according to the first historical data and the second historical data; A strategy generation module, used to establish a first judgment strategy based on the first efficiency model and the second rainfall impact model; The judgment module is used to obtain the first real-time data and the second real-time data of the target photovoltaic panel, and automatically clean the target photovoltaic panel in combination with the first judgment strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images