Dust accumulation amount prediction method and device, storage medium and electronic device

Abstract

The invention discloses a dust accumulation amount prediction method and device, a storage medium and an electronic device.The method comprises the steps that dust data of a photovoltaic panel are collected according to a preset frequency, and the dust data at least comprise meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment; the dust data are input into a first model to predict the dust accumulation amount of the photovoltaic panel at the first moment, the first model is a pre-trained recurrent neural network model used for predicting the dust accumulation amount, and the first moment is the next moment of the current moment; a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency. The problem that the prediction effect of quantitative prediction of dust accumulation of the photovoltaic panel is poor is solved, and the technical effect of accurate prediction of the dust accumulation amount is achieved.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic power generation, and in particular to a method and device for predicting dust accumulation, a storage medium, and an electronic device. Background Art

[0002] In the context of the increasingly severe global energy crisis, the gradual depletion of traditional fossil energy and the serious pollution problems it causes to the environment have promoted the attention and investment of various countries in renewable energy. As an important renewable energy, solar energy has become one of the important directions for the development and utilization of new energy with its inexhaustible, clean and environmentally friendly characteristics. Photovoltaic power generation technology is a technology that directly converts solar energy into electrical energy using the photovoltaic effect, and its core device is the photovoltaic panel. Photovoltaic panels are usually installed in outdoor environments with sufficient light to maximize the absorption of sunlight. However, in actual applications, photovoltaic panels, especially when installed in deserts and other windy and sandy areas, will be affected by dust, sand and other particles. These particles are deposited on the surface of photovoltaic panels to form a dust accumulation layer, which seriously affects the light transmittance of photovoltaic panels, thereby reducing their photoelectric conversion efficiency.

[0003] Dust accumulation not only leads to a decrease in the power generation efficiency of photovoltaic panels, but also accelerates the aging and damage of photovoltaic panels, affecting their service life. Studies have shown that dust accumulation will cause the output power loss of photovoltaic systems to be about 2% to 10%, and in extreme cases, the output power loss can even reach 25%. This not only increases the operating cost of photovoltaic power stations, but also reduces the economic benefits of photovoltaic power generation. To solve this problem, photovoltaic power stations need to clean photovoltaic panels regularly. However, the cleaning work of most photovoltaic power stations currently relies mainly on manual experience and lacks a scientific cleaning plan. This not only leads to unreasonable cleaning frequency and timing, but may also cause increased cleaning costs and waste of resources.

[0004] Dust accumulation prediction is an important basis for formulating scientific cleaning plans. Accurate dust accumulation prediction can help photovoltaic power stations reasonably arrange cleaning time and frequency, thereby reducing the impact of dust on photovoltaic panel performance and improving the power generation efficiency and economic benefits of photovoltaic systems. At present, most of the research on dust accumulation of photovoltaic panels focuses on qualitative analysis and experimental research, and lacks an effective quantitative prediction model. Traditional dust accumulation prediction methods are mainly based on empirical formulas and statistical analysis. These methods have certain limitations when dealing with complex nonlinear relationships and multivariate coupling effects, and it is difficult to accurately predict the dynamic changes of dust accumulation.

[0005] There is no effective solution to the problem of poor prediction effect of quantitative prediction of dust accumulation on photovoltaic panels in the related art. Summary of the invention

[0006] The embodiments of the present application provide a method and device for predicting dust accumulation, a storage medium, and an electronic device to at least solve the problem of poor prediction effect of quantitative prediction of dust accumulation of photovoltaic panels in the prior art.

[0007] According to one embodiment of the embodiments of the present application, a method for predicting the amount of dust accumulation is provided, comprising: collecting dust data of a photovoltaic panel at a preset frequency, wherein the dust data includes at least: meteorological data of an environment where the photovoltaic panel is located, and air quality data of the environment; inputting the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting the amount of dust accumulation, the first moment is the next moment of the current moment, and a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency.

[0008] In an exemplary embodiment, before the dust data is input into the first model to predict the dust accumulation amount of the photovoltaic panel at the first moment, the method also includes: optimizing the model parameters of the second model by a pigeon flock optimization algorithm to obtain a third model, wherein the second model is an untrained recursive neural network model; training the third model by historical data of the photovoltaic panel collected from the experimental platform to obtain the first model, wherein the historical data includes: historical dust accumulation amounts of the photovoltaic panel corresponding to multiple second moments, historical meteorological data and historical air quality data corresponding to the historical dust accumulation amounts, and a third time interval between two adjacent second moments among the multiple second moments is the same as the second time interval.

[0009] In an exemplary embodiment, the model parameters of the second model are optimized by a pigeon flock optimization algorithm to obtain a third model, including: determining the initial position and speed of the model parameters by optimization factors; updating the initial position and speed by update rules to obtain a first updated position and an updated speed; determining the optimized model parameters by the first updated position and the updated speed to obtain the third model.

[0010] In an exemplary embodiment, the optimized model parameters are determined by the first update position and the update speed to obtain the third model, including: optimizing the first update position by Cauchy distribution to obtain the second update position; and determining the optimized model parameters by the second update position and the update speed to obtain the third model.

[0011] In an exemplary embodiment, the dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, including: inputting the dust data into the first model to obtain a first output result, wherein the first output result at least includes: a deposition rate of particulate matter in the air; and inputting the first output result into a target formula to obtain the dust accumulation amount.

[0012] In an exemplary embodiment, after the dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, the method further includes: inputting the amount of dust accumulation into a regression model to obtain a second output result; and determining the second output result as the power loss of the photovoltaic panel.

[0013] According to another embodiment of the embodiments of the present application, a dust accumulation prediction device is also provided, including: a collection module, used to collect dust data of a photovoltaic panel at a preset frequency, wherein the dust data includes at least: meteorological data of the environment where the photovoltaic panel is located, and air quality data of the environment; an input module, used to input the dust data into a first model to predict the dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency.

[0014] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above method when running.

[0015] According to another aspect of the embodiments of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method through the computer program.

[0016] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0017] In an embodiment of the present application, dust data of the photovoltaic panel is collected at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment; the dust data is input into a first model to predict the amount of dust accumulated on the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting the amount of dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency. Through the above embodiment, the problem of poor prediction effect of quantitative prediction of dust accumulation of photovoltaic panels in the prior art is solved, and the technical effect of accurate prediction of dust accumulation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for predicting dust accumulation amount according to an embodiment of the present application;

[0020] Figure 2 is a flow chart of a method for predicting dust accumulation according to an embodiment of the present application;

[0021] Figure 3 is another flow chart of a method for predicting dust accumulation amount according to an embodiment of the present application;

[0022] Figure 4 is a structural schematic diagram of a first model of a method for predicting dust accumulation amount according to an embodiment of the present application;

[0023] Figure 5 4 is a structural block diagram of a device for predicting dust accumulation amount according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices; "plurality" means two or more.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or similar computing device or cloud platform or independent physical server or software platform, wherein the above software platform is run through one or more servers. Taking running on a computer terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal of a method for predicting dust accumulation in an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 and memory 104 for storing data. In an exemplary embodiment, the computer terminal may also include a transmission device 106 for communication functions and an input and output device 108, wherein the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. A person skilled in the art can understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Equivalent functions or comparisons shown Figure 1 A different configuration with more features is shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as computer programs corresponding to the methods in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, that is, the above-mentioned methods are implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] In this embodiment, a method for predicting the amount of dust accumulation is provided, which is applied to the above-mentioned computer terminal. Figure 2 : is a flow chart of a method for predicting dust accumulation according to an embodiment of the present application, the process comprising the following steps:

[0030] Step S202, collecting dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment;

[0031] Step S204: input the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency.

[0032] Through the above device, the dust data of the photovoltaic panel is collected at a preset frequency, wherein the dust data at least includes: the meteorological data of the environment where the photovoltaic panel is located, and the air quality data of the environment; the dust data is input into the first model to predict the dust accumulation of the photovoltaic panel at the first moment, wherein the first model is a pre-trained recursive neural network model for predicting the dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency. Through the above embodiment, the problem of poor prediction effect of quantitative prediction of dust accumulation of photovoltaic panels in the prior art is solved, and the technical effect of accurate prediction of dust accumulation is achieved.

[0033] In some optional embodiments, before the dust data is input into the first model to predict the amount of dust accumulation on the photovoltaic panel at the first moment, the first model needs to be acquired. Figure 3 As shown, the step of obtaining the first model includes:

[0034] Step S302, optimizing the model parameters of the second model by a pigeon flock optimization algorithm to obtain a third model, wherein the second model is an untrained recursive neural network model;

[0035] Step S304, training the third model by using the historical data of the photovoltaic panels collected from the experimental platform to obtain the first model, wherein the historical data includes: historical dust accumulation of the photovoltaic panels corresponding to multiple second moments, historical meteorological data and historical air quality data corresponding to the historical dust accumulation, and a third time interval between two adjacent second moments among the multiple second moments is the same as the second time interval.

[0036] It is understandable that an untrained recursive neural network model is first determined, for example, a Delay-Output Echo State Network (DESN) model. Secondly, the model parameters of the untrained recursive neural network model are optimized by a pigeon flock optimization algorithm to obtain a third model. Finally, the third model is trained by the historical data of the photovoltaic panels collected from the experimental platform to obtain the first model.

[0037] Therefore, the first model obtained in the embodiment of the present application is a model trained with historical data, and the model parameters are optimized by the pigeon flock optimization algorithm, which can more accurately predict the amount of dust accumulation.

[0038] In an exemplary embodiment, the model parameters of the second model are optimized by a pigeon flock optimization algorithm to obtain a third model, including: determining the initial position and speed of the model parameters by optimization factors; updating the initial position and speed by update rules to obtain a first updated position and an updated speed; determining the optimized model parameters by the first updated position and the updated speed to obtain the third model.

[0039] Optionally, the optimization factor is a map and compass factor. The model parameters are used as "pigeons" in the pigeon flock optimization algorithm, and the initial position and speed of each pigeon are determined by the optimization factor, and the initial position and speed are updated.

[0040] Furthermore, the optimized model parameters are determined by the first update position and the update speed to obtain the third model, including: optimizing the first update position by Cauchy distribution to obtain the second update position; and determining the optimized model parameters by the second update position and the update speed to obtain the third model.

[0041] That is, the adaptive Cauchy mutation strategy is used to further optimize the position of the pigeon group to enhance the diversity and convergence speed of the search. This is specifically achieved through the density function of the Cauchy distribution.

[0042] In an exemplary embodiment, the dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, including: inputting the dust data into the first model to obtain a first output result, wherein the first output result at least includes: a deposition rate of particulate matter in the air; and inputting the first output result into a target formula to obtain the dust accumulation amount.

[0043] Optionally, the target formula is as follows:

[0044] m=∫(v PM10 C PM10 +v PM2.5 C PM2.5 )tcosθdt; Formula 1

[0045] Among them, v PM10 and v PM2.5 are the deposition rates of PM10 and PM2.5 particles in the air, C PM10 and C PM2.5 is the corresponding concentration of particulate matter in the air, t is the time step, and θ is the installation angle of the photovoltaic panel.

[0046] In an exemplary embodiment, after the dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, the method further includes: inputting the amount of dust accumulation into a regression model to obtain a second output result; and determining the second output result as the power loss of the photovoltaic panel.

[0047] Optionally, the regression model can refer to Formula 2:

[0048] η dust =A+β1X1+β2X2+β3X3+β4X4+β5X5+β6X6; Formula 2

[0049] Among them, η dust represents power loss, A is the regression coefficient, X1 is the temperature of the photovoltaic panel, X2 is the dust concentration, X3 is the humidity, X4 is the irradiance, X5 is the rainfall, and X6 is the ambient temperature. β1, ..., β6 all represent weights.

[0050] In order to better understand the process of the above-mentioned method for predicting the amount of dust accumulation, the above-mentioned method for predicting the amount of dust accumulation is described below in combination with an optional embodiment, but it is not used to limit the technical solution of the embodiment of the present application.

[0051] With the development of artificial intelligence technology, prediction methods based on artificial neural networks (ANN) have shown their unique advantages in dealing with complex nonlinear problems. As a special recurrent neural network (RNN), the echo state network (ESN) has a simple structure, efficient training and powerful time series prediction capabilities. It has received widespread attention in the field of time series prediction in recent years. ESN captures the dynamic characteristics of data by mapping the input data into a high-dimensional, sparsely connected dynamic reservoir and then reading the output linearly. It is suitable for modeling and prediction of complex nonlinear systems. However, there are still some challenges in the practical application of ESN. For example, the selection of reservoir parameters (including leakage rate, spectral radius, input scaling, etc.) has a great impact on model performance. How to effectively optimize these parameters has become the key to improving the prediction performance of ESN.

[0052] Based on the characteristics of ESN, the optional embodiment of the present application proposes a photovoltaic panel dust accumulation daily prediction method based on delayed output echo state network (DESN), combined with the pigeon inspired optimization (PIO) algorithm of adaptive Cauchy (AC) mutation strategy, and optimizes the reservoir parameters of DESN to improve the accuracy and reliability of the prediction model. This method uses meteorological and air quality data to establish a dust accumulation prediction model, which provides a scientific basis for the cleaning strategy and grid dispatch of photovoltaic power stations, and has important practical application value and broad application prospects.

[0053] The specific implementation steps of the optional embodiment of this application are as follows:

[0054] First, a delayed output echo state network (DESN) model is established (equivalent to the second model in the above embodiment). The echo state network is a special recursive neural network, such as Figure 4 As shown, its structure consists of an input layer, a reservoir and an output layer. In an optional embodiment of the present application, the reservoir state update and output equations of the DESN are defined as Formula 3 and Formula 4 respectively:

[0055] x(n+1)=(1-a)x(n)+f(W in u(n+1)+Wx(n)+W fb1 y(n)+W fb2 y(n-τ)); Formula 3

[0056] y(n)=g(W out [x(n);u(n)])-αy(n-τ); Formula 4

[0057] Where a is the leakage rate, α is the delay factor, τ is the delay range, and W in , W and W fb1 , W fb2 are the input weight matrix, the reservoir weight matrix and the output feedback weight matrix, respectively. out is the output weight matrix.

[0058] Among them, u = u(n), x = x(n) and y = y(n) represent the external input vector, reservoir state and output vector respectively.

[0059] in, Minimizing the error e(n) is used to train the output weight W of the network out , so that the predicted output y(n) is consistent with the expected output Match.

[0060] Next, the key parameters of the DESN model are optimized using the pigeon swarm optimization algorithm (PIO). In the PIO algorithm, the initial position and speed of each pigeon are first determined by the map and compass factors, and the update rules are as follows:

[0061] V i (n) = V i (n-1)*e -Rn +rand*(X gbest -X i (n-1)); Formula 5

[0062] X i (n) = X i (n-1)+V i (n); Formula 6

[0063] Among them, R is the map and compass factor, rand is a random number, and X gbest is the global optimal position, V i represents the speed of the pigeon, X i Indicates the position of the pigeon.

[0064] On this basis, the adaptive Cauchy mutation strategy is used to further optimize the position of the pigeon flock to enhance the diversity and convergence speed of the search. The density function of the Cauchy distribution is defined as Formula 7:

[0065]

[0066] Where x0 represents the pigeon position of the input Cauchy distribution density function.

[0067] By performing Cauchy distribution perturbations on the pigeon positions in the PIO algorithm, the DESN model parameters (such as leakage rate a and spectrum radius γ) are optimized. The goal is to minimize the root mean square error (RMSE). The calculation formula is shown in Formula 8:

[0068]

[0069] Then, the optimized DESN model (equivalent to the third model) is trained using the dust accumulation on the surface of the photovoltaic panels, meteorological data (such as humidity, irradiance) and air quality data (such as PM2.5 and PM10 concentrations) collected by the experimental platform. During the training process, considering the time series characteristics of the input data and the lag effect of dust accumulation, historical data (i.e., data such as the dust accumulation on the surface of the photovoltaic panels collected by the above-mentioned experimental platform) are used as input to train the model to predict the dust accumulation at future time points.

[0070] After the model training is completed, the model (i.e., the first model) is applied to the prediction of actual data. According to the inclination angle of the photovoltaic panel, the PM2.5 and PM10 concentrations, humidity, irradiance and other parameters in the environment, the optimized DESN model (i.e., the first model) is used to predict the daily dust accumulation. The calculation formula of the dust accumulation is shown in the above formula 1.

[0071] Finally, the predicted dust accumulation can be used to calculate the power loss of the photovoltaic system, thus providing a scientific basis for the cleaning and maintenance plan of the photovoltaic power station. The specific power loss calculation can be done using a regression model, see formula 2.

[0072] In summary, the optional embodiments of the present application achieve high-precision prediction of the dust accumulation condition (i.e., the amount of dust accumulation) of photovoltaic panels, and can provide strong data support for the maintenance and cleaning strategies of photovoltaic systems.

[0073] The optional embodiment of the present application improves the accuracy of dust accumulation prediction by optimizing reservoir parameters (including leakage rate, spectral radius and input scaling) by combining the echo state network (DESN) of delayed output with the pigeon flock optimization algorithm (AC-PIO) of adaptive Cauchy mutation strategy. Using meteorological and air quality data (such as humidity, wind speed, wind direction, PM2.5, PM10 and rainfall) as input variables, the dust accumulation data recorded by the experimental platform are used for model training and testing, the relationship between environmental parameters and dust accumulation is analyzed, and a dust accumulation prediction model based on AC-PIO-DESN is constructed. This model not only improves the accuracy and reliability of prediction, but also effectively solves the limitations of traditional methods when dealing with complex nonlinear relationships and multivariate coupling effects. By comparing experimental results, it is proved that the optional embodiment of the present application is superior in meteorological and air quality data prediction, and has high application value. The technical points to be protected include the structural design of the delayed output echo state network, the adaptive optimization method of reservoir parameters and the dust accumulation prediction method based on these technologies. It can provide a scientific basis for dust cleaning strategies and grid dispatching of photovoltaic power stations, significantly improve the power generation efficiency and economic benefits of photovoltaic systems, and has important practical application value and broad application prospects.

[0074] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0075] The present application also provides a structural block diagram of a device for predicting dust accumulation. Figure 5 is a structural block diagram of a device for predicting dust accumulation according to an embodiment of the present application; Figure 5 As shown, including:

[0076] The collection module 52 is used to collect dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment;

[0077] An input module 54 is used to input the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency.

[0078] Through the above device, the dust data of the photovoltaic panel is collected at a preset frequency, wherein the dust data at least includes: the meteorological data of the environment where the photovoltaic panel is located, and the air quality data of the environment; the dust data is input into the first model to predict the dust accumulation of the photovoltaic panel at the first moment, wherein the first model is a pre-trained recursive neural network model for predicting the dust accumulation, the first moment is the next moment of the current moment, and the first time interval between the current moment and the first moment is the same as the second time interval corresponding to the preset frequency. Through the above embodiment, the problem of poor prediction effect of quantitative prediction of dust accumulation of photovoltaic panels in the prior art is solved, and the technical effect of accurate prediction of dust accumulation is achieved.

[0079] In an exemplary embodiment, before the dust data is input into the first model to predict the dust accumulation amount of the photovoltaic panel at the first moment, the device also includes a training module, which is used to: optimize the model parameters of the second model by a pigeon flock optimization algorithm to obtain a third model, wherein the second model is an untrained recursive neural network model; train the third model by using historical data of the photovoltaic panel collected from the experimental platform to obtain the first model, wherein the historical data includes: historical dust accumulation amounts of the photovoltaic panel corresponding to multiple second moments, historical meteorological data and historical air quality data corresponding to the historical dust accumulation amounts, and the third time interval between two adjacent second moments among the multiple second moments is the same as the second time interval.

[0080] In an exemplary embodiment, the training module is also used to: determine the initial position and speed of the model parameters through optimization factors; update the initial position and speed through update rules to obtain a first updated position and an updated speed; determine the optimized model parameters through the first updated position and the updated speed to obtain the third model.

[0081] In an exemplary embodiment, the training module is further used to: optimize the first update position through Cauchy distribution to obtain the second update position; determine the optimized model parameters through the second update position and the update speed to obtain the third model.

[0082] In an exemplary embodiment, the input module 54 is also used to: input the dust data into the first model to obtain a first output result, wherein the first output result at least includes: the deposition rate of particulate matter in the air; input the first output result into the target formula to obtain the dust accumulation amount.

[0083] In an exemplary embodiment, after the dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, the device also includes a determination module for: inputting the amount of dust accumulation into a regression model to obtain a second output result; and determining the second output result as the power loss of the photovoltaic panel.

[0084] An embodiment of the present application further provides a storage medium, which includes a stored program, wherein the program executes any of the above methods when it is run.

[0085] Optionally, in this embodiment, the storage medium may be configured to store program codes for executing the following steps:

[0086] S1, collecting dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment;

[0087] S2, inputting the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency.

[0088] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0089] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0090] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0091] S1, collecting dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment;

[0092] S2, inputting the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency.

[0093] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0094] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0095] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0096] An embodiment of the present application also provides a computer program, which includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in any one of the above method embodiments.

[0097] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0098] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0099] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting dust accumulation, characterized in that: include: Collect dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located and air quality data of the environment; The dust data is input into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency.

2. The method for predicting dust accumulation according to claim 1, characterized in that: Before inputting the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, the method further comprises: Optimizing the model parameters of the second model by a pigeon flock optimization algorithm to obtain a third model, wherein the second model is an untrained recursive neural network model; The first model is obtained by training the third model with historical data of the photovoltaic panels collected from the experimental platform, wherein the historical data includes: historical dust accumulation of the photovoltaic panels corresponding to multiple second moments, historical meteorological data and historical air quality data corresponding to the historical dust accumulation, and a third time interval between two adjacent second moments among the multiple second moments is the same as the second time interval.

3. The method for predicting dust accumulation according to claim 2, characterized in that: The model parameters of the second model are optimized by the pigeon flock optimization algorithm to obtain the third model, including: Determining the initial position and velocity of the model parameters by optimizing the factors; The initial position and speed are updated by an update rule to obtain a first updated position and an updated speed; The optimized model parameters are determined by the first update position and the update speed to obtain the third model.

4. The method for predicting dust accumulation according to claim 3, characterized in that: Determining optimized model parameters by using the first update position and the update speed to obtain the third model includes: Optimizing the first update position by using Cauchy distribution to obtain a second update position; The optimized model parameters are determined by the second update position and the update speed to obtain the third model.

5. The method for predicting dust accumulation according to claim 1, characterized in that: Inputting the dust data into a first model to predict the amount of dust accumulation on the photovoltaic panel at a first moment comprises: Inputting the dust data into the first model to obtain a first output result, wherein the first output result at least includes: a deposition rate of particulate matter in the air; The first output result is input into a target formula to obtain the dust accumulation amount.

6. The method for predicting dust accumulation according to claim 1, characterized in that: After inputting the dust data into a first model to predict the amount of dust accumulation on the photovoltaic panel at a first moment, the method further comprises: Inputting the dust accumulation amount into a regression model to obtain a second output result; The second output result is determined as the power loss of the photovoltaic panel.

7. A device for predicting dust accumulation, characterized in that: include: A collection module, used for collecting dust data of the photovoltaic panel at a preset frequency, wherein the dust data at least includes: meteorological data of the environment where the photovoltaic panel is located, and air quality data of the environment; An input module is used to input the dust data into a first model to predict the amount of dust accumulation of the photovoltaic panel at a first moment, wherein the first model is a pre-trained recursive neural network model for predicting dust accumulation, the first moment is the next moment of the current moment, and a first time interval between the current moment and the first moment is the same as a second time interval corresponding to the preset frequency.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product, comprising a computer program, 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 6 are implemented.

Images