Control method and system for photovoltaic panel cleaning

Abstract

The invention belongs to the technical field of photovoltaic power generation, and particularly relates to a control method and system for photovoltaic panel cleaning. Comprising a data acquisition module, an image acquisition module, a data processing module, an image recognition unit, a stain component analysis module, a cleaning strategy making module, an equipment control module, an effect evaluation module, a power generation efficiency monitoring module, a cleaning decision module and a display module. Accurate dirt judgment and intelligent cleaning strategy making are achieved, the cleaning efficiency and the energy-saving effect are improved, the labor cost is reduced, the service life of equipment is prolonged, the power generation efficiency is improved, the system can adapt to complex environments and stain changes, the defects of a traditional cleaning mode are effectively overcome, and development of the photovoltaic industry is promoted.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation, and in particular to a control method and system for cleaning a photovoltaic panel. Background Art

[0002] In the current energy field, solar photovoltaic power generation has been widely used and developed as a clean and sustainable way of energy utilization; however, photovoltaic panels will inevitably be contaminated by various dirt during long-term exposure to the outdoor environment, such as dust, bird droppings, oil and air pollutants. The accumulation of these dirt will significantly reduce the power generation efficiency of photovoltaic panels and seriously affect the overall performance and economic benefits of photovoltaic power generation systems.

[0003] Traditional photovoltaic panel cleaning methods mainly include manual cleaning and timed mechanical cleaning; manual cleaning is not only labor-intensive and inefficient, but also limited by factors such as labor costs and operational safety, making it difficult to be widely used in large-scale photovoltaic power stations; although timed mechanical cleaning has improved the cleaning efficiency to a certain extent, it often lacks accurate judgment of the actual dirt condition and environmental conditions of the photovoltaic panels, and there are problems of excessive cleaning or untimely cleaning; excessive cleaning will lead to unnecessary energy consumption and equipment wear, while untimely cleaning will not be able to fully tap the power generation potential of the photovoltaic panels, resulting in energy waste.

[0004] In addition, the existing cleaning methods are relatively simple in formulating cleaning strategies, usually using fixed cleaning methods, cleaning paths and cleaning parameters, and cannot be flexibly adjusted according to different dirt types, dirt levels and real-time environmental factors (such as temperature, humidity, wind speed, etc.); this makes it impossible to effectively control the energy consumption during the cleaning process, and it is also difficult to ensure that the best cleaning effect can be achieved under various complex working conditions, which cannot meet the urgent needs of modern photovoltaic power generation systems for efficient, energy-saving and intelligent cleaning.

[0005] In summary, developing a control method and system for cleaning photovoltaic panels that can accurately determine the dirt status of photovoltaic panels and intelligently optimize cleaning strategies to improve cleaning efficiency and energy-saving effects has become a technical problem that needs to be urgently solved in this field. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, the technical solution adopted by the present invention to solve the technical problem is: a control method for cleaning a photovoltaic panel, comprising the following steps:

[0007] S1: Use sensors to collect real-time power generation P(t), light intensity I(t), surface dirt level S(t), temperature T(t), humidity H(t) and wind speed V(t) data of the photovoltaic panel, and obtain image information of the photovoltaic panel surface through the image acquisition module;

[0008] S2: Analyze the collected data based on a preset data analysis model to determine the dirt condition on the surface of the photovoltaic panel, wherein the dirt condition includes the type of dirt and the proportion of the dirt coverage area, wherein the collected image is analyzed and processed by an image recognition unit to identify the type of dirt, and a comprehensive judgment is made in combination with the sensor data;

[0009] S3: Through the stain component analysis module, the external database is connected via the Internet to query and obtain stain component information according to the identified stain type;

[0010] S4: Based on the dirt judgment result, the stain composition information and the current environmental conditions, and in combination with the preset cleaning strategy library, an intelligent optimization algorithm is used to formulate a cleaning strategy, wherein the intelligent optimization algorithm is used to optimize the cleaning efficiency η and the energy saving efficiency E save As the goal, by continuously iterating various parameters in the cleaning strategy, the energy consumption and cleaning time in the cleaning process are minimized while ensuring that the cleaning effect meets the predetermined standards. The calculation method of the cleaning efficiency η is Where P after is the power generation of the photovoltaic panel after cleaning, P before is the power generation of the photovoltaic panel before cleaning, E clean The energy consumed during the cleaning process includes the energy consumption of cleaning equipment and the energy consumption of cleaning agents. The energy consumption of cleaning equipment and the equipment operating power P eq (t) and running time t clean Related, the energy consumption of detergent use and the amount of detergent used Q chem And unit cleaning agent energy consumption coefficient K chem Related;

[0011] S5: converting the cleaning strategy into a control instruction and sending it to the cleaning device to drive the cleaning device to perform a cleaning operation on the photovoltaic panel according to a predetermined cleaning path and parameters;

[0012] S6: After the cleaning is completed, the power generation and surface condition data of the photovoltaic panel are collected again and compared with the data before cleaning to evaluate the effect of this cleaning;

[0013] S7: If the cleaning effect does not meet the expected goal, the cleaning strategy will be adjusted and optimized based on the evaluation results, and the data and experience of this cleaning process will be recorded to improve the cleaning strategy library and data analysis model.

[0014] Furthermore, the energy saving effect E save The calculation method is E save =E normal -E actual , where E normal is the energy consumed by conventional cleaning methods, E actualis the energy actually consumed in this cleaning; the specific steps adopted by the intelligent optimization algorithm include: initializing the population of cleaning strategy parameters, each individual represents a set of cleaning strategy parameters; calculating the cleaning efficiency η and energy saving efficiency E corresponding to each individual save , calculate the fitness of each individual according to the fitness function; use selection, crossover and mutation operations to evolve the population and generate a new generation of individuals; repeat the above steps until the preset number of iterations is met or the convergence condition is reached, and select the individual with the highest fitness as the optimal cleaning strategy.

[0015] Furthermore, the sensors include a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor; the image acquisition module includes a high-definition camera, whose installation position and shooting angle can cover the main surface area of ​​the photovoltaic panel to obtain a clear and complete surface image, and the camera has automatic focus, dimming and image enhancement functions to adapt to different lighting and environmental conditions.

[0016] A control system for cleaning a photovoltaic panel, comprising:

[0017] A data acquisition module, composed of a plurality of sensors, for collecting operation data and environmental data of the photovoltaic panel in real time and transmitting the data to the data processing module, wherein the plurality of sensors include a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor, and also includes an image acquisition module for acquiring an image of the surface of the photovoltaic panel;

[0018] A data processing module is used to receive and store the data collected by the data collection module, analyze and process the data using a built-in data analysis model, determine the dirt status of the photovoltaic panel, and send the processing results to the cleaning strategy formulation module;

[0019] An image recognition unit is connected to the image acquisition module, analyzes and processes the acquired image to identify the type of stains on the surface of the photovoltaic panel, and transmits the result to the data processing module;

[0020] The stain component analysis module connects to the external database via the Internet, queries and obtains stain component information based on the identified stain types, and transmits the information to the cleaning strategy formulation module;

[0021] A cleaning strategy formulation module is used to generate a cleaning strategy using an intelligent optimization algorithm based on the received dirt judgment results, stain composition information and a preset cleaning strategy library, including a cleaning method, a cleaning path, a cleaning time, operating parameters of a cleaning device and the selection of a cleaning agent, and send the cleaning strategy to the device control module. The intelligent optimization algorithm aims to optimize cleaning efficiency and energy saving effects;

[0022] An equipment control module is used to receive the cleaning strategy sent by the cleaning strategy formulation module and convert it into a control instruction, drive the cleaning equipment to perform cleaning operations according to predetermined parameters and paths, and monitor the operating status of the cleaning equipment in real time;

[0023] The effect evaluation module is used to evaluate the cleaning effect of the photovoltaic panels after cleaning is completed, adjust and optimize the cleaning strategy according to the evaluation results, and feed back the evaluation data to the data processing module for updating the data analysis model and the cleaning strategy library;

[0024] The power generation efficiency monitoring module is used to record the power generation efficiency curve of the photovoltaic panel and transmit the data to the cleaning decision module;

[0025] A cleaning decision module is connected to the power generation efficiency monitoring module and the image recognition unit, and determines whether cleaning is needed according to the stain condition and the power generation efficiency curve. When the effect of the stain on the photovoltaic power generation efficiency reaches a preset threshold, the cleaning strategy formulation module is triggered to generate a cleaning strategy;

[0026] The display module is connected to the cleaning strategy formulation module and is used to display specific cleaning methods, cleaning agent selection and cleaning equipment operating parameter information, including water pressure value, water volume, etc. on a mobile phone or computer.

[0027] Furthermore, the intelligent optimization algorithm is an improved algorithm based on any one of the genetic algorithm, particle swarm optimization algorithm or simulated annealing algorithm, and its optimization objective function is a comprehensive indicator of cleaning efficiency and energy-saving effect. The optimal solution is sought by continuously adjusting the cleaning strategy parameters to achieve efficient cleaning and energy saving and consumption reduction.

[0028] Furthermore, the image recognition unit adopts an image recognition algorithm based on deep learning and is pre-trained with a variety of stain image samples. The training samples include images of different types of stains under different lighting, angles and surface materials to improve the accuracy and robustness of stain type identification.

[0029] Furthermore, the external database connected to the Internet by the stain component analysis module includes a professional chemical substance database, an environmental pollutant database and a stain analysis knowledge base, ensuring that the acquired stain component information is accurate and reliable, and can be updated according to the latest scientific research results and actual stain analysis data.

[0030] Furthermore, the cleaning strategy library is classified, stored and updated according to different regions, different seasons, different photovoltaic panel types and different dirt conditions. The stored cleaning strategies are verified and optimized through actual cleaning effects to adapt to diverse cleaning needs, and are combined with intelligent optimization algorithms to continuously improve cleaning efficiency and energy-saving effects.

[0031] Furthermore, it also includes a weather information interaction module, which is connected to an external weather system to obtain weather information for the current and future period of time, including precipitation probability, temperature changes, etc., and transmits the weather information to the data processing module and the display module.

[0032] Furthermore, it also includes a stain change prediction module, which is connected to the stain component analysis module and the weather information interaction module, and predicts the change of stains over time based on the stain components, the current ambient temperature and humidity, and the acquired weather information. When the stain components will cause corrosion to the surface of the photovoltaic panel, or the stain solidifies and has strong adhesion over time and is easy to clean in the early stage but difficult to clean in the later stage, a reminder message is sent to the display module to show how long it will take for the stain to become more difficult to clean under the current weather conditions, and adjust the time report according to the weather changes; the stain change prediction module makes predictions based on a relationship model between stain characteristics, ambient temperature and humidity and time. The model is obtained by fitting and optimizing experimental data of different stains under various environmental conditions, and can be adaptively adjusted according to actual stain changes.

[0033] The beneficial effects of the present invention are as follows:

[0034] 1. The control method and system for cleaning photovoltaic panels described in the present invention can accurately judge the dirt condition of photovoltaic panels (including dirt type, coverage area ratio and stain composition, etc.) by collecting rich data through various sensors (such as power, light, dirt, temperature, humidity, wind speed sensors and image acquisition modules) and combining intelligent optimization algorithms, and formulate personalized cleaning strategies accordingly; this avoids the problems of excessive cleaning or untimely cleaning in traditional cleaning methods, and significantly improves the cleaning efficiency while ensuring that the cleaning effect meets the predetermined standards, while minimizing the energy consumption in the cleaning process, achieving high efficiency and energy saving, and effectively improving the overall performance and economic benefits of the photovoltaic power generation system.

[0035] 2. The photovoltaic panel cleaning control method and system described in the present invention can dynamically adjust the cleaning strategy and schedule according to real-time environmental conditions (such as precipitation probability, temperature change, humidity, etc.) and the characteristics of stain changes over time through the cooperation between the weather information interaction module and the stain change prediction module. For stains that are easily affected by the environment and change (such as stains with enhanced adhesion after solidification at a specific temperature and humidity), early warning can be given and cleaning operations can be reasonably planned to prevent stains from being difficult to clean due to long-term accumulation or environmental factors, thereby reducing the work intensity and wear of the cleaning equipment, extending the service life of the equipment, reducing the equipment maintenance cost, and ensuring that the photovoltaic panel always maintains a good power generation state in various complex environments.

[0036] 3. The control method and system for cleaning photovoltaic panels described in the present invention achieves high intelligence and automation in the entire cleaning process through the accurate identification of stain types by the image recognition unit based on deep learning, the precise query of stain components by the stain component analysis module, and the continuous iterative optimization of cleaning strategies by the intelligent optimization algorithm; the cleaning strategy library is classified and updated according to different regions, seasons, photovoltaic panel types, and dirt conditions, further enhancing the system's adaptability to diverse cleaning needs. This intelligent cleaning system not only improves the accuracy and reliability of cleaning operations, but also reduces manual intervention, reduces labor costs and operational risks, provides strong support for the efficient operation of large-scale photovoltaic power stations, and promotes the intelligent development of the solar photovoltaic power generation industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention will be further described below in conjunction with the accompanying drawings.

[0038] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0039] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0040] like Figure 1 As shown, the present invention proposes a control method for cleaning a photovoltaic panel, comprising the following steps:

[0041] S1: Use sensors to collect real-time power generation power P(t), light intensity I(t), surface dirt level S(t), temperature T(t), humidity H(t) and wind speed V(t) data of the photovoltaic panel, and use the image acquisition module to obtain image information on the surface of the photovoltaic panel; multiple sensors and image acquisition modules synchronously collect data and capture images of the photovoltaic panel and its surrounding environment according to the established sampling frequency and working sequence.

[0042] When working, the power sensor converts the generated power into electrical signals through electromagnetic induction and other principles, the light sensor senses the light intensity based on the photoelectric effect, the dirt sensor detects the degree of dirt using optical, acoustic or electrical means, the temperature, humidity and wind speed sensors use the thermal, hygroscopic and mechanical principles to obtain the corresponding environmental data respectively, and the image acquisition module focuses the surface image of the photovoltaic panel on the image sensor and converts it into a digital signal through the optical imaging principle; these data and image information are quickly and stably transmitted to the data processing module through wired or wireless transmission channels, providing comprehensive and real-time raw data for subsequent analysis;

[0043] Through the collaborative operation of various types of sensors and high-precision image acquisition modules, all-round and real-time monitoring of the operating status and surface dirt conditions of photovoltaic panels is achieved, overcoming the one-sidedness and lag of a single data source; the rich data obtained lays a solid foundation for accurately judging the dirt conditions and formulating scientific cleaning strategies, effectively improving the accuracy of the assessment of photovoltaic panel cleaning needs, and thus providing a strong guarantee for improving photovoltaic power generation efficiency and reducing power generation losses caused by dirt. At the same time, it enhances the reliability and adaptability of the entire cleaning control system, enabling it to cope with photovoltaic panel cleaning tasks under different environmental conditions, reducing the uncertainty of cleaning effects caused by environmental changes, extending the effective power generation time and service life of photovoltaic panels, and optimizing the performance and economic benefits of photovoltaic power generation systems as a whole.

[0044] S2: Analyze the collected data based on a preset data analysis model to determine the dirt condition on the surface of the photovoltaic panel, wherein the dirt condition includes the type of dirt and the proportion of the dirt coverage area, wherein the collected image is analyzed and processed by an image recognition unit to identify the type of stain, and a comprehensive judgment is made in combination with the sensor data.

[0045] When working, after receiving the sensor and image data, the data processing module first pre-processes them, including denoising, normalization and other operations to optimize the data quality; then, the processed data is input into the data analysis model built based on machine learning algorithms (such as support vector machines, neural networks, etc.). The model establishes a mapping relationship between data features and dirt conditions by learning a large amount of historical data, so as to determine the type of dirt and the coverage area ratio; at the same time, the image recognition unit uses deep learning image recognition algorithms (such as convolutional neural networks (CNN)) to analyze the collected images; first, the image is pre-processed by grayscale and enhancement, and then it is input into the trained CNN model. The model compares the stains with the pre-stored sample library based on the color, texture, shape and other features of the stains, identifies the type of stains, and feeds the results back to the data processing module. Finally, the data processing module combines other sensor data for comprehensive analysis to obtain accurate dirt condition information.

[0046] The dirt judgment method that combines data analysis with image recognition greatly improves the accuracy and efficiency of dirt condition judgment, and has higher precision and reliability than traditional manual judgment or simple sensor threshold judgment. Accurate dirt judgment can provide key decision-making basis for the formulation of subsequent cleaning strategies, avoiding the waste of cleaning resources and the problems of untimely or excessive cleaning. By targeted selection of cleaning methods, cleaning agents and equipment operating parameters, not only the cleaning effect is improved, so that photovoltaic panels can quickly restore high power generation efficiency, but also the cleaning cost, including manpower, material and time costs, is reduced. At the same time, the potential damage to photovoltaic panels caused by improper cleaning operations is reduced, the service life of photovoltaic panels is extended, the stability and sustainability of photovoltaic power generation systems are improved, and strong support is provided for the efficient operation and maintenance of large-scale photovoltaic power stations.

[0047] S3: Through the stain component analysis module, the Internet is used to connect to the external database, and the stain component information is queried and obtained according to the identified stain type.

[0048] When working, after the image recognition unit determines the type of stain, the stain component analysis module immediately starts the Internet connection program and sends a query request containing the type of stain to a professional external database (covering a chemical substance database, an environmental pollutant database, and a stain analysis knowledge base, etc.); the database quickly searches and matches according to the request, and uses its rich storage resources and powerful retrieval function to find the corresponding detailed information of the stain components, including chemical composition, structure, reaction characteristics, and environmental change trends, and transmits this information back to the stain component analysis module, which then organizes and quickly transmits it to the cleaning strategy formulation module. The entire process is efficient and fast, ensuring the timeliness and accuracy of the information required for cleaning strategy formulation;

[0049] By using the Internet to connect to external professional databases to obtain information on stain composition, the system's depth and breadth of understanding of stains has been significantly expanded, and it can formulate accurate and effective cleaning plans for complex and diverse stains. This not only improves cleaning efficiency, ensures that the ideal effect can be achieved with one cleaning, and reduces the number and cost of repeated cleaning, but also reduces the amount of detergent used and potential pollution to the environment by selecting the most suitable detergents and cleaning methods, thereby improving the environmental friendliness of the cleaning process. At the same time, accurate stain composition analysis can effectively protect the surface materials of photovoltaic panels, avoid corrosion, scratches and other damage to photovoltaic panels due to improper chemical reactions between detergents and stains or excessive cleaning, extend the service life of photovoltaic panels, ensure the long-term and stable operation of photovoltaic power generation systems, and improve the performance and economic benefits of photovoltaic panel cleaning systems from multiple aspects, thereby enhancing their competitiveness and practicality in the market.

[0050] S4: Based on the dirt judgment result, the stain composition information and the current environmental conditions, and in combination with the preset cleaning strategy library, an intelligent optimization algorithm is used to formulate a cleaning strategy, wherein the intelligent optimization algorithm is used to optimize the cleaning efficiency η and the energy saving efficiency E save As the goal, by continuously iterating various parameters in the cleaning strategy, the energy consumption and cleaning time in the cleaning process are minimized while ensuring that the cleaning effect meets the predetermined standards. The calculation method of the cleaning efficiency η is Where P after is the power generation of the photovoltaic panel after cleaning, P before is the power generation of the photovoltaic panel before cleaning, E clean The energy consumed during the cleaning process includes the energy consumption of cleaning equipment and the energy consumption of cleaning agents. The energy consumption of cleaning equipment and the equipment operating power P eq (t) and running time t clean Related, the energy consumption of detergent use and the amount of detergent used Q chem And unit cleaning agent energy consumption coefficient K chem Related.

[0051] When working, after obtaining the dirt judgment results, stain composition information and environmental condition data, the cleaning strategy formulation module extracts the initial cleaning strategy parameter combination from the preset cleaning strategy library that is classified and stored according to different regions, seasons, photovoltaic panel types and dirt conditions and has been optimized through actual verification, so as to initialize the population of the intelligent optimization algorithm (such as improved genetic algorithm, particle swarm optimization algorithm or simulated annealing algorithm, etc.); then, the cleaning efficiency η and energy saving effect E corresponding to each individual are calculated. save The fitness of individuals is evaluated according to the fitness function; then, the population is evolved through selection, crossover and mutation operations. The selection operation uses methods such as tournament selection to select excellent individuals to enter the next generation. The crossover operation exchanges the cleaning strategy parameters between individuals with a certain probability. The mutation operation randomly perturbs some parameters to generate a new generation of individuals. Repeat these steps until the preset number of iterations or convergence conditions are met. Finally, the individual with the highest fitness is selected as the optimal cleaning strategy, covering cleaning methods (dry cleaning, water washing, etc.), cleaning paths, cleaning time and equipment operating parameters (water pressure, brush speed, etc.);

[0052] The cleaning strategy is formulated by combining intelligent optimization algorithms with a cleaning strategy library, which fully integrates multiple factors and achieves dual optimization of cleaning efficiency and energy-saving effects. Through precise parameter iteration and optimization, the power generation of cleaned photovoltaic panels is greatly improved compared to uncleaned photovoltaic panels, while significantly reducing energy consumption and time consumption during the cleaning process, saving energy resources such as water and electricity and manpower costs. The intelligent optimization strategy formulation method is highly adaptable and flexible, and can automatically adjust the cleaning plan according to different working conditions to ensure efficient, energy-saving, and high-quality cleaning effects in various complex environments and dirt conditions, effectively improving the economic and environmental benefits of photovoltaic power generation systems. Moreover, with continuous practice and data accumulation, the cleaning strategy library and algorithms can continuously improve and optimize themselves, further improve system performance, and provide solid and reliable technical support for the intelligent and sustainable operation and maintenance of large-scale photovoltaic power stations, promoting technological progress and development in the photovoltaic power generation industry.

[0053] S5: converting the cleaning strategy into a control instruction and sending it to the cleaning device to drive the cleaning device to perform a cleaning operation on the photovoltaic panel according to a predetermined cleaning path and parameters.

[0054] During operation, after receiving the cleaning strategy generated by the cleaning strategy formulation module, the equipment control module quickly parses and processes it, converts information such as cleaning method, cleaning path, cleaning time and equipment operating parameters into control command signals that can be recognized and executed by the cleaning equipment, and encodes and transmits them according to the control interface and communication protocol of the cleaning equipment (such as industrial Ethernet, Bluetooth or Wi-Fi, etc.); for high-pressure water guns, sends instructions including start, stop, water flow pressure adjustment, spray gun movement speed and angle control; for brush cleaning devices, sends brush motor start, stop, speed adjustment and brush lifting and moving control instructions; cleaning equipment After receiving the instruction, the controller drives the corresponding actuator (motor, water pump, valve, etc.) to act according to the instruction, so that the cleaning equipment cleans the photovoltaic panels in an orderly manner along the predetermined cleaning path (through programmed path planning or real-time path tracking based on sensors), and adjusts the equipment operation parameters in real time according to the instruction to ensure the cleaning process is stable and efficient; during the cleaning process, the equipment control module also collects the equipment operation status data in real time through the sensors installed on the equipment (motor current, water pressure, position sensor, etc.), and feeds it back to the data processing module for analysis and processing, so as to timely discover and handle equipment failures or abnormal conditions, and ensure the smooth progress of the cleaning operation;

[0055] The cleaning strategy is accurately converted into equipment control instructions through the equipment control module, achieving highly automated and intelligent control of the cleaning process, greatly improving the accuracy, stability and efficiency of the cleaning operation; compared with traditional manual or simple timing controlled cleaning equipment, this precise control ensures that the cleaning equipment operates in the best condition, effectively avoiding problems such as poor cleaning effect, equipment damage and energy waste caused by human operating errors or unreasonable parameter settings; real-time equipment operation status monitoring and feedback mechanism further enhances the safety and reliability of the cleaning process, can promptly discover and solve problems in the cleaning process, reduce equipment failure rate and maintenance costs, ensure the stable operation of the photovoltaic power generation system, and improve the performance and benefits of the entire photovoltaic panel cleaning system, providing a strong guarantee for the efficient operation of the photovoltaic power station, while also providing a more convenient and safe working environment for operators, reducing labor intensity and operational risks.

[0056] S6: After the cleaning is completed, the power generation and surface condition data of the photovoltaic panel are collected again and compared with the data before cleaning to evaluate the effect of this cleaning.

[0057] During operation, after cleaning is completed, the data acquisition module quickly collects the power generation and surface condition data of the photovoltaic panel again in the same collection method and parameter settings as before cleaning, and transmits these data to the effect evaluation module through a stable transmission channel; after receiving the data, the effect evaluation module immediately calculates the difference between the power generation data before and after cleaning, and obtains the power improvement ratio, such as the formula: At the same time, image analysis and dirt detection technology are used to conduct in-depth analysis of the surface condition data after cleaning to determine whether the residual dirt ratio is lower than the predetermined threshold. Through the comprehensive evaluation of these two key indicators, it is quickly and accurately determined whether the cleaning effect has achieved the expected goal. The entire evaluation process is efficient and rigorous, and clear results can be obtained in a short time after cleaning is completed, providing timely basis for subsequent decision-making;

[0058] By comparing the data before and after cleaning to evaluate the cleaning effect, a quantitative assessment of the cleaning quality is achieved, which is objective and scientific; accurate evaluation results can promptly discover problems and deficiencies in the cleaning process, and provide strong data support for further optimization of cleaning strategies; if the cleaning effect does not meet expectations, the cleaning method, time, equipment parameters or cleaning agents can be adjusted in a targeted manner, and then the cleaning operation can be performed again until a satisfactory effect is achieved, forming a closed-loop optimization control mechanism; this mechanism continuously improves the performance and cleaning quality of the photovoltaic panel cleaning system, ensures that the photovoltaic panels always maintain a high power generation efficiency, maximizes the power generation potential of the photovoltaic power generation system, and improves the economic benefits and reliability of photovoltaic power generation; at the same time, by continuously optimizing the cleaning strategy, unnecessary cleaning operations and resource waste are reduced, the service life of photovoltaic panels is extended, the total cost of photovoltaic power generation is reduced, the competitiveness of photovoltaic power generation in the energy market is enhanced, and the sustainable development of the clean energy industry is promoted.

[0059] S7: If the cleaning effect does not meet the expected goal, the cleaning strategy will be adjusted and optimized based on the evaluation results, and the data and experience of this cleaning process will be recorded to improve the cleaning strategy library and data analysis model.

[0060] During operation, when the effect evaluation module determines that the cleaning effect does not meet expectations, the cleaning strategy will be adjusted and optimized immediately according to the problems such as insufficient power increase and excessive dirt residue in the evaluation results. For example, if it is found that there is serious dirt residue in a certain area, the cleaning time of the area may be increased, the water flow pressure may be increased, or a more effective cleaning agent may be replaced. If the power generation increase does not reach the target, the cleaning method may be adjusted or the cleaning path may be optimized. At the same time, all data in this cleaning process, including the collected sensor data, image data, cleaning strategy adopted, cleaning equipment operating parameters, and final cleaning effect evaluation data and other detailed information, will be recorded and summarized and analyzed in combination with actual experience. These data and experience will be fed back to the data processing module to further improve the parameter settings and feature extraction methods in the data analysis model, so that it can more accurately judge the dirt condition and predict the cleaning effect. At the same time, the optimized cleaning strategy will be updated to the cleaning strategy library to enrich and optimize the content of the strategy library, so that when encountering similar dirt conditions in the future, effective cleaning strategies can be formulated faster and more accurately, and the ability of the entire system to cope with various complex cleaning tasks will be continuously improved.

[0061] The feedback adjustment mechanism based on cleaning effect evaluation enables the photovoltaic panel cleaning system to have the ability of self-learning and self-improvement; by continuously recording and analyzing data and experience in the cleaning process, the system can gradually adapt to different regions, different seasons, different photovoltaic panel types and complex and changeable dirt conditions, continuously optimize cleaning strategies and data analysis models, and improve the stability and reliability of cleaning effects; this will not only help improve the power generation efficiency and service life of current photovoltaic panels and reduce operation and maintenance costs, but also as the system continues to learn and evolve, its application value in the entire photovoltaic power generation industry will continue to increase, and it will be able to provide more photovoltaic power stations with efficient, intelligent and sustainable cleaning solutions, promote photovoltaic power generation technology to develop in a more efficient, stable and environmentally friendly direction, and make positive contributions to the optimization and sustainable development of the global energy structure.

[0062] As a specific implementation of the present invention, the energy saving effect E save The calculation method is E save =E normal -E actual , where E normal is the energy consumed by conventional cleaning methods, E actual is the energy actually consumed in this cleaning; the specific steps adopted by the intelligent optimization algorithm include: initializing the population of cleaning strategy parameters, each individual represents a set of cleaning strategy parameters; calculating the cleaning efficiency η and energy saving efficiency E corresponding to each individual save , calculate the fitness of each individual according to the fitness function; use selection, crossover and mutation operations to evolve the population and generate a new generation of individuals; repeat the above steps until the preset number of iterations is met or the convergence condition is reached, and select the individual with the highest fitness as the optimal cleaning strategy.

[0063] When working, by accurately calculating the energy-saving effect, the advantage of this method in energy consumption compared with traditional cleaning methods can be intuitively demonstrated, which is helpful to significantly reduce energy consumption costs in large-scale photovoltaic power station cleaning operations, improve energy utilization efficiency, and comply with the development trend of energy conservation and environmental protection; the iterative process of the intelligent optimization algorithm can continuously explore better cleaning strategy parameter combinations, further improve the balance between cleaning efficiency and energy-saving effects, and minimize the waste of energy and resources while ensuring the quality of cleaning, thereby enhancing the competitiveness and practicality of the present invention in the market, providing key technical support for the sustainable development of the photovoltaic industry, and promoting the industry's technological progress and the realization of energy conservation and emission reduction goals.

[0064] As a specific embodiment of the present invention, the sensor includes a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor; the image acquisition module includes a high-definition camera, whose installation position and shooting angle can cover the main surface area of ​​the photovoltaic panel to obtain a clear and complete surface image, and the camera has automatic focus, dimming and image enhancement functions to adapt to different lighting and environmental conditions.

[0065] When working, various types of sensors work together to obtain comprehensive and real-time information on the operating status and surrounding environment of the photovoltaic panels, providing a rich data basis for accurately judging the dirt condition and improving the accuracy and reliability of dirt detection; the high-definition camera and its advanced functions ensure that the image details of the photovoltaic panel surface can be clearly captured, regardless of the lighting and environmental conditions, it can provide high-quality image data for the image recognition unit, so as to more accurately identify the type of stains, further optimize the cleaning strategy, improve the cleaning effect, reduce the problem of incomplete or excessive cleaning due to misjudgment of dirt, extend the service life of photovoltaic panels, improve the performance and stability of photovoltaic power generation systems, reduce operation and maintenance costs, and enhance the adaptability and robustness of the entire system.

[0066] A control system for cleaning a photovoltaic panel, comprising:

[0067] A data acquisition module, composed of a plurality of sensors, for collecting operation data and environmental data of the photovoltaic panel in real time and transmitting the data to the data processing module, wherein the plurality of sensors include a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor, and also includes an image acquisition module for acquiring an image of the surface of the photovoltaic panel;

[0068] A data processing module is used to receive and store the data collected by the data collection module, analyze and process the data using a built-in data analysis model, determine the dirt status of the photovoltaic panel, and send the processing results to the cleaning strategy formulation module;

[0069] An image recognition unit is connected to the image acquisition module, analyzes and processes the acquired image to identify the type of stains on the surface of the photovoltaic panel, and transmits the result to the data processing module;

[0070] The stain component analysis module connects to the external database via the Internet, queries and obtains stain component information based on the identified stain types, and transmits the information to the cleaning strategy formulation module;

[0071] A cleaning strategy formulation module is used to generate a cleaning strategy using an intelligent optimization algorithm based on the received dirt judgment results, stain composition information and a preset cleaning strategy library, including a cleaning method, a cleaning path, a cleaning time, operating parameters of a cleaning device and the selection of a cleaning agent, and send the cleaning strategy to the device control module. The intelligent optimization algorithm aims to optimize cleaning efficiency and energy saving effects;

[0072] An equipment control module is used to receive the cleaning strategy sent by the cleaning strategy formulation module and convert it into a control instruction, drive the cleaning equipment to perform cleaning operations according to predetermined parameters and paths, and monitor the operating status of the cleaning equipment in real time;

[0073] The effect evaluation module is used to evaluate the cleaning effect of the photovoltaic panels after cleaning is completed, adjust and optimize the cleaning strategy according to the evaluation results, and feed back the evaluation data to the data processing module for updating the data analysis model and the cleaning strategy library;

[0074] The power generation efficiency monitoring module is used to record the power generation efficiency curve of the photovoltaic panel and transmit the data to the cleaning decision module;

[0075] A cleaning decision module is connected to the power generation efficiency monitoring module and the image recognition unit, and determines whether cleaning is needed according to the stain condition and the power generation efficiency curve. When the effect of the stain on the photovoltaic power generation efficiency reaches a preset threshold, the cleaning strategy formulation module is triggered to generate a cleaning strategy;

[0076] The display module is connected to the cleaning strategy formulation module and is used to display specific cleaning methods, cleaning agent selection and cleaning equipment operating parameter information, including water pressure value, water volume, etc. on a mobile phone or computer.

[0077] During operation, through close collaboration and efficient data interaction between modules, full automation and intelligent management of photovoltaic panel cleaning are achieved; the comprehensiveness and accuracy of the data acquisition module provide a solid foundation for subsequent dirt judgment and strategy formulation, reduce the interference of human factors, and improve the scientificity and timeliness of cleaning decisions; the application of intelligent optimization algorithm in the cleaning strategy formulation module can quickly generate the optimal cleaning plan according to different working conditions, significantly reduce energy consumption while improving cleaning efficiency, and improve the economy and environmental protection of the entire system; the feedback mechanism of the effect evaluation module enables the system to continuously learn and improve, adapt to various complex and changeable dirt conditions and environmental conditions, and ensure the stability and reliability of the cleaning effect; the display module facilitates users to monitor and manage the cleaning process in real time, improves the system's ease of use and user experience; effectively improves the cleaning quality and efficiency of photovoltaic panels, reduces operation and maintenance costs, improves the power generation efficiency and stability of photovoltaic power generation systems, and promotes the photovoltaic power generation industry to develop in the direction of intelligence and efficiency, with broad market application prospects and important economic and environmental value.

[0078] As a specific implementation of the present invention, the intelligent optimization algorithm is an improved algorithm based on any one of the genetic algorithm, particle swarm optimization algorithm or simulated annealing algorithm, and its optimization objective function is a comprehensive indicator of cleaning efficiency and energy-saving effect. The optimal solution is sought by continuously adjusting the cleaning strategy parameters to achieve efficient cleaning and energy saving and consumption reduction.

[0079] When working, an improved version of the intelligent optimization algorithm is used, which can give full play to its advantages in solving complex problems, explore the parameter space of the cleaning strategy more efficiently, and find the optimal balance between cleaning efficiency and energy saving effect; compared with traditional optimization algorithms or cleaning strategies with fixed parameters, the present invention can quickly and accurately adjust the cleaning method, path, time and equipment parameters according to factors such as the actual dirt condition of the photovoltaic panel, environmental conditions and equipment performance, so as to achieve precise cleaning and avoid waste of resources and unstable cleaning effect; this not only improves the power generation efficiency of the photovoltaic panel and reduces energy consumption, but also extends the service life of the photovoltaic panel and reduces maintenance costs, providing reliable guarantees for the long-term stable operation of the photovoltaic power generation system, enhancing the adaptability and competitiveness of the system in different scenarios, and helping to promote the technological upgrading and sustainable development of the photovoltaic industry, and has significant social and economic benefits in terms of energy conservation, emission reduction and improving energy utilization efficiency.

[0080] As a specific embodiment of the present invention, the image recognition unit adopts an image recognition algorithm based on deep learning and is pre-trained with a variety of stain image samples. The training samples include images of different types of stains under different lighting, angles and surface materials to improve the accuracy and robustness of stain type identification.

[0081] When working, the image recognition algorithm based on deep learning has powerful feature extraction and pattern recognition capabilities. By training a large number of diverse stain image samples, it can accurately identify the types of stains in various complex situations, and is not affected by changes in lighting, shooting angles, and different surface materials of photovoltaic panels. This enables the system to more accurately understand the dirt situation on the surface of photovoltaic panels, and provide more detailed and accurate information for the subsequent formulation of cleaning strategies, so as to select more suitable cleaning agents and cleaning methods, improve the cleaning effect, and reduce the risk of incomplete cleaning or damage to photovoltaic panels due to incorrect stain recognition. At the same time, high recognition accuracy and robustness reduce the system's dependence on human intervention, improve the degree of automation and efficiency of the cleaning process, and further improve the performance and reliability of the entire photovoltaic panel cleaning control system, providing strong support for the efficient operation and maintenance of large-scale photovoltaic power stations, which will help improve the economic benefits and competitiveness of the photovoltaic power generation industry and promote the intelligent development of the industry.

[0082] As a specific embodiment of the present invention, the external database connected to the Internet by the stain component analysis module includes a professional chemical substance database, an environmental pollutant database and a stain analysis knowledge base, ensuring that the acquired stain component information is accurate and reliable, and can be updated according to the latest scientific research results and actual stain analysis data.

[0083] When working, connecting to an external database greatly enriches the system's cognitive scope and depth of stain components, and can obtain the most accurate and detailed stain component information, including its chemical composition, properties, possible reaction characteristics, and changing trends under different environmental conditions; this enables the cleaning strategy formulation module to select the most suitable detergent and cleaning method according to the specific components of the stain, avoiding poor cleaning effects or damage to photovoltaic panels due to improper selection of detergents, and improving the pertinence and effectiveness of cleaning; at the same time, the timely update capability of the database ensures that the system can master the latest stain analysis knowledge and technology, adapt to the ever-changing stain types and environmental conditions, and always stay at the forefront of stain treatment technology, further improving the performance and reliability of the entire system, providing a solid guarantee for the long-term and efficient cleaning of photovoltaic panels, helping to extend the service life of photovoltaic panels, reduce the cost of photovoltaic power generation, improve the overall benefits of photovoltaic power generation systems, promote the sustainable development of the photovoltaic power generation industry, and enhance the competitiveness and adaptability of the present invention in the market.

[0084] As a specific implementation of the present invention, the cleaning strategy library is classified, stored and updated according to different regions, different seasons, different photovoltaic panel types and different dirt conditions. The stored cleaning strategies are verified and optimized through actual cleaning effects to adapt to diverse cleaning needs, and are combined with intelligent optimization algorithms to continuously improve cleaning efficiency and energy-saving effects.

[0085] When working, the classified, stored and updated cleaning strategy library has strong practicality and adaptability. It can quickly provide initial cleaning strategies verified by practice according to the climate characteristics of different regions, the dirt change rules in different seasons, the material and structural characteristics of different photovoltaic panel types, and various complex dirt conditions, providing a high-quality initial solution space for the intelligent optimization algorithm, greatly shortening the time for finding the optimal cleaning strategy; combined with the intelligent optimization algorithm, it can further optimize the cleaning strategy on the basis of existing experience, continuously improve the cleaning efficiency and energy-saving effect, and meet diverse cleaning needs; this enables the present invention to be widely used in various photovoltaic power stations around the world, effectively improving the cleaning quality and efficiency of photovoltaic panels, reducing cleaning costs and energy consumption, extending the service life of photovoltaic panels, and improving the overall performance and economic benefits of photovoltaic power generation systems, promoting the standardization, intelligence and sustainable development of the photovoltaic power generation industry, and has important industrial value and social significance.

[0086] As a specific implementation of the present invention, it also includes a weather information interaction module, which is connected to an external weather system to obtain weather information for the current and future period of time, including precipitation probability, temperature changes, etc., and transmits the weather information to the data processing module and the display module.

[0087] When working, the existence of the weather information interaction module enables the system to obtain weather changes in real time, and predict in advance the impact of changes in environmental factors such as precipitation and temperature on the dirt condition and cleaning operations of photovoltaic panels; when the probability of precipitation is high, the system can adjust the cleaning plan in advance to avoid unnecessary cleaning operations in the rain, or use rainwater to naturally clean certain easily washed dirt, saving cleaning resources and costs; when the temperature changes greatly, it can reasonably adjust the concentration of detergents and the operating parameters of cleaning equipment according to the impact of temperature on stain characteristics and cleaning effects, to ensure the stability and reliability of the cleaning effect; at the same time, the weather information is displayed on the display module, which is convenient for operators to understand the weather conditions in a timely manner and make more reasonable decisions, thereby improving the intelligence level of the system and user experience, and further enhancing the adaptability and practicality of the entire photovoltaic panel cleaning control system, providing a strong guarantee for the stable operation of the photovoltaic power generation system, helping to improve the efficiency and economic benefits of photovoltaic power generation, and promoting the development of the clean energy industry.

[0088] As a specific embodiment of the present invention, it also includes a stain change prediction module, which is connected to the stain component analysis module and the weather information interaction module. According to the stain components, the current ambient temperature and humidity, and the acquired weather information, it predicts the change of the stain over time. When the stain components will cause corrosion to the surface of the photovoltaic panel, or the stain solidifies and has strong adhesion over time and is easy to clean in the early stage but difficult to clean in the later stage, a reminder message is sent to the display module to show how long it will take for the stain to become more difficult to clean under the current weather conditions, and adjust the time report according to the weather changes. The stain change prediction module makes predictions based on a relationship model between stain characteristics, ambient temperature and humidity, and time. The model is obtained by fitting and optimizing experimental data of different stains under various environmental conditions, and can be adaptively adjusted according to actual stain changes.

[0089] When working, the stain change prediction module can predict the change trend of stains in advance by comprehensively considering the stain composition, environmental temperature and humidity, and weather information, providing a more forward-looking basis for cleaning decisions; for stains that may cause corrosion to photovoltaic panels or change over time and are difficult to clean, reminder information is sent in time, so that operators can arrange cleaning operations in advance to avoid irreversible damage to photovoltaic panels caused by stain solidification or corrosion, thereby extending the service life of photovoltaic panels; the report of stain change time is dynamically adjusted according to weather changes, ensuring the accuracy and timeliness of the prediction, further optimizing the formulation of cleaning plans, improving the utilization efficiency of cleaning resources, and reducing cleaning costs and maintenance difficulties; this intelligent stain change prediction function enhances the preventive maintenance capability of the entire photovoltaic panel cleaning control system, improves the reliability and stability of the system, helps to improve the long-term performance and economic benefits of photovoltaic power generation systems, and promotes the photovoltaic power generation industry to develop in a more refined and intelligent direction, and has important application value and innovative significance in the field of clean energy.

[0090] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A control method for cleaning a photovoltaic panel, characterized in that: The steps include: S1: Use sensors to collect real-time power generation P(t), light intensity I(t), surface dirt level S(t), temperature T(t), humidity H(t) and wind speed V(t) data of the photovoltaic panel, and obtain image information of the photovoltaic panel surface through the image acquisition module; S2: Analyze the collected data based on a preset data analysis model to determine the dirt condition on the surface of the photovoltaic panel, where the dirt condition includes the type of dirt and the proportion of the dirt coverage area; S3: Through the stain component analysis module, the external database is connected via the Internet to query and obtain stain component information according to the identified stain type; S4: Based on the dirt judgment result, the stain composition information and the current environmental conditions, and in combination with the preset cleaning strategy library, an intelligent optimization algorithm is used to formulate a cleaning strategy, wherein the intelligent optimization algorithm is used to optimize the cleaning efficiency η and the energy saving efficiency E save As the goal, by continuously iterating various parameters in the cleaning strategy, the energy consumption and cleaning time in the cleaning process are minimized while ensuring that the cleaning effect meets the predetermined standards. The calculation method of the cleaning efficiency η is Where P after is the power generation of the photovoltaic panel after cleaning, P before is the power generation of the photovoltaic panel before cleaning, E clean The energy consumed during the cleaning process includes the energy consumption of cleaning equipment and the energy consumption of cleaning agents. The energy consumption of cleaning equipment and the equipment operating power P eq (t) and running time t clean Related, the energy consumption of detergent and the amount of detergent used Q chem And unit cleaning agent energy consumption coefficient K chem Related; S5: converting the cleaning strategy into a control instruction and sending it to the cleaning device to drive the cleaning device to perform a cleaning operation on the photovoltaic panel according to a predetermined cleaning path and parameters; S6: After the cleaning is completed, the power generation and surface condition data of the photovoltaic panel are collected again and compared with the data before cleaning to evaluate the effect of this cleaning; S7: If the cleaning effect does not meet the expected goal, the cleaning strategy will be adjusted and optimized based on the evaluation results, and the data and experience of this cleaning process will be recorded to improve the cleaning strategy library and data analysis model.

2. A photovoltaic panel cleaning control method according to claim 1, characterized in that: The energy saving effect E save The calculation method is E save =E normal -E actual , where E normal is the energy consumed by conventional cleaning methods, E actual The actual energy consumed for this cleaning; The specific steps adopted by the intelligent optimization algorithm include: initializing a population of cleaning strategy parameters, each individual representing a set of cleaning strategy parameters; Calculate the cleaning efficiency η and energy saving efficiency E corresponding to each individual save , calculate the fitness of each individual according to the fitness function; use selection, crossover and mutation operations to evolve the population and generate a new generation of individuals; repeat the above steps until the preset number of iterations is met or the convergence condition is reached, and select the individual with the highest fitness as the optimal cleaning strategy.

3. A photovoltaic panel cleaning control method according to claim 1, characterized in that: The sensors include a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor; the image acquisition module includes a high-definition camera, the installation position and shooting angle of which can cover the main surface area of ​​the photovoltaic panel to obtain a clear and complete surface image.

4. A photovoltaic panel cleaning control system, applicable to the photovoltaic panel cleaning control method according to any one of claims 1 to 3, characterized in that: include: A data acquisition module, composed of a plurality of sensors, for collecting operation data and environmental data of the photovoltaic panel in real time and transmitting the data to the data processing module, wherein the plurality of sensors include a power sensor, a light sensor, a dirt sensor, a temperature sensor, a humidity sensor and a wind speed sensor, and also includes an image acquisition module for acquiring an image of the surface of the photovoltaic panel; A data processing module is used to receive and store the data collected by the data collection module, analyze and process the data using a built-in data analysis model, determine the dirt status of the photovoltaic panel, and send the processing results to the cleaning strategy formulation module; An image recognition unit is connected to the image acquisition module, analyzes and processes the acquired image to identify the type of stains on the surface of the photovoltaic panel, and transmits the result to the data processing module; The stain component analysis module connects to the external database via the Internet, queries and obtains stain component information based on the identified stain types, and transmits the information to the cleaning strategy formulation module; A cleaning strategy formulation module is used to generate a cleaning strategy using an intelligent optimization algorithm based on the received dirt judgment results, stain composition information and a preset cleaning strategy library, including a cleaning method, a cleaning path, a cleaning time, operating parameters of a cleaning device and the selection of a cleaning agent, and send the cleaning strategy to the device control module. The intelligent optimization algorithm aims to optimize cleaning efficiency and energy saving effects; An equipment control module is used to receive the cleaning strategy sent by the cleaning strategy formulation module and convert it into a control instruction, drive the cleaning equipment to perform cleaning operations according to predetermined parameters and paths, and monitor the operating status of the cleaning equipment in real time; The effect evaluation module is used to evaluate the cleaning effect of the photovoltaic panels after cleaning is completed, adjust and optimize the cleaning strategy according to the evaluation results, and feed back the evaluation data to the data processing module for updating the data analysis model and the cleaning strategy library; The power generation efficiency monitoring module is used to record the power generation efficiency curve of the photovoltaic panel and transmit the data to the cleaning decision module; A cleaning decision module is connected to the power generation efficiency monitoring module and the image recognition unit, and determines whether cleaning is needed according to the stain condition and the power generation efficiency curve. When the effect of the stain on the photovoltaic power generation efficiency reaches a preset threshold, the cleaning strategy formulation module is triggered to generate a cleaning strategy; The display module is connected to the cleaning strategy formulation module and is used to display specific cleaning methods, cleaning agent selection and cleaning equipment operating parameter information on a mobile phone or computer.

5. The photovoltaic panel cleaning control system according to claim 4, characterized in that: The intelligent optimization algorithm is an improved algorithm based on any one of a genetic algorithm, a particle swarm optimization algorithm or a simulated annealing algorithm, and its optimization objective function is a comprehensive index of cleaning efficiency and energy-saving effect.

6. The photovoltaic panel cleaning control system according to claim 4, characterized in that: The image recognition unit adopts an image recognition algorithm based on deep learning and is pre-trained with a variety of stain image samples. The training samples include images of different types of stains under different lighting, angles and surface materials.

7. The photovoltaic panel cleaning control system according to claim 4, characterized in that: The external databases connected to the Internet by the stain component analysis module include a professional chemical substance database, an environmental pollutant database and a stain analysis knowledge base.

8. The photovoltaic panel cleaning control system according to claim 4, characterized in that: The cleaning strategy library is classified, stored and updated according to different regions, different seasons, different photovoltaic panel types and different dirt conditions.

9. The photovoltaic panel cleaning control system according to claim 4, characterized in that: It also includes a weather information interaction module, which is connected to an external weather system to obtain current and future weather information and transmit the weather information to the data processing module and the display module.

10. The photovoltaic panel cleaning control system according to claim 9, characterized in that: It also includes a stain change prediction module, which is connected to the stain composition analysis module and the weather information interaction module, and predicts the change of stains over time based on the stain composition, current ambient temperature and humidity, and acquired weather information; the stain change prediction module makes predictions based on a relationship model between stain characteristics, ambient temperature and humidity, and time.

Images