Intelligent self-adaptive cleaning system and method for photovoltaic module

Abstract

The invention discloses an intelligent self-adaptive cleaning system and method for a photovoltaic module, and mainly solves the problems that the cleaning effect of a photovoltaic module cleaning system on stains (such as bird droppings and oil stains) with strong adhesive force and fine dust particles is limited, and it is difficult to ensure that the surface of the module keeps good cleanliness all the time. The cleaning system comprises an environment sensing module used for detecting the cleaning state of the photovoltaic module, an intelligent cleaning module connected with the environment sensing module and used for executing the cleaning function, and self-adaptive control connected with the environment sensing module and used for controlling the intelligent cleaning module. And the energy recovery module is connected with the intelligent cleaning module and is used for realizing energy recovery and feeding back power supply to the environment sensing module. By means of the design, the multi-sensor fusion environment sensing technology and the intelligent cleaning technology are combined, the stain type and the pollution degree can be more accurately recognized, a targeted cleaning mode is adopted, and compared with a traditional cleaning technology, the cleaning effect is remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation, and in particular, relates to an intelligent adaptive cleaning system and method for photovoltaic components. Background Art

[0002] In the field of photovoltaic power generation, there are many problems with the existing methods for cleaning photovoltaic modules. Manual cleaning is labor-intensive and costly. It is difficult to ensure the timeliness and consistency of cleaning due to the constraints of personnel working time and efficiency, which makes it easy for dust and other pollutants to accumulate on the surface of the module for a long time, affecting the power generation efficiency. Mechanical automatic cleaning devices also have many shortcomings. Their structure is single and it is difficult to adapt to changes in different environmental conditions and pollution conditions. In complex climate environments such as strong wind and sand, high humidity or frequent bird activities, they cannot effectively deal with various pollution conditions. Moreover, their cleaning methods may cause scratches and damage to the surface of the module, affecting the service life and power generation performance of the module. At the same time, the cleaning effect of stains with strong adhesion (such as bird droppings, oil stains, etc.) and fine dust particles is limited, and it is difficult to ensure that the surface of the module always maintains good cleanliness. This is mainly due to technical limitations. The sensor application of existing cleaning technology is not comprehensive and accurate. It only relies on simple humidity or light sensors to judge the cleaning time, and cannot accurately perceive various factors such as dust concentration, wind direction and wind speed, resulting in unreasonable cleaning strategies. At the same time, the design concept is limited, focusing on a single cleaning method, not integrating multiple effective means, lacking intelligent adaptive control, and cannot flexibly adjust cleaning parameters and methods.

[0003] Therefore, there is an urgent need for an intelligent adaptive cleaning system and method for photovoltaic components as proposed in the present invention to address the shortcomings of existing cleaning methods, improve the cleaning efficiency and effect of photovoltaic components, ensure power generation efficiency, adapt to different environmental conditions, and achieve intelligent, efficient, reliable and environmentally friendly cleaning and maintenance. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent adaptive cleaning system and method for photovoltaic modules, mainly to solve the problem that the cleaning system of photovoltaic modules has limited cleaning effect on stains with strong adhesion (such as bird droppings, oil stains, etc.) and fine dust particles, and it is difficult to ensure that the surface of the module always maintains good cleanliness.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A photovoltaic module intelligent adaptive cleaning system includes an environmental perception module for detecting the cleaning status of the photovoltaic module, an intelligent cleaning module connected to the environmental perception module for performing a cleaning function, an adaptive control connected to the environmental perception module for controlling the intelligent cleaning module, and an energy recovery module connected to the intelligent cleaning module for realizing energy recovery and feeding back power to the environmental perception module.

[0007] Furthermore, in the present invention, the intelligent cleaning module includes an ultrasonic vibration cleaning subsystem, an intelligent jet cleaning subsystem, a flexible rotating cleaning brush system, an ultrasonic transducer, a negative pressure suction device and a nozzle installed on the photovoltaic module; wherein the ultrasonic vibration cleaning subsystem realizes ultrasonic vibration cleaning through an ultrasonic transducer; the image recognition sensor identifies stains, the intelligent jet cleaning subsystem cleans the stains by spraying detergent through a nozzle, and recovers residual substances using a negative pressure suction device, and finally wipes and cleans using a flexible rotating cleaning brush system.

[0008] Furthermore, in the present invention, the environment perception module includes

[0009] Dust concentration sensor: uses the principle of laser scattering to accurately measure the concentration of dust particles in the air near the component surface;

[0010] Humidity sensor: real-time monitoring of ambient humidity;

[0011] Rainfall sensor: accurately detects whether there is rainfall and the amount of rainfall;

[0012] Wind direction and speed sensor: obtain wind direction and wind speed information;

[0013] Image recognition sensor: It captures component surface images and processes them using a convolutional neural network.

[0014] Based on the above system, the present invention also provides a photovoltaic module intelligent adaptive cleaning method, comprising:

[0015] S1, using the environment sensing module to detect the cleanliness status of the photovoltaic module;

[0016] S2, based on the detection results of the environmental perception module, the adaptive control module uses an intelligent algorithm based on fuzzy logic and neural network to establish a cleaning decision model and generate the optimal cleaning strategy in real time;

[0017] S3, according to the cleaning strategy of the adaptive control module, start the intelligent cleaning module to clean the photovoltaic module;

[0018] S4, recovering the energy generated during the cleaning process through an energy recovery module.

[0019] Furthermore, in step S1, the working process of the environment perception module is as follows:

[0020] S11, the dust concentration sensor uses the principle of laser scattering to measure the concentration of dust particles in the air near the component surface, calculates the dust concentration based on information such as the intensity and angle of the scattered light, and outputs the data;

[0021] S12, the humidity sensor monitors the ambient humidity in real time and outputs humidity data;

[0022] S13, the rainfall sensor accurately detects whether there is rainfall and the amount of rainfall, and outputs rainfall information data;

[0023] S14, a wind direction and wind speed sensor obtains wind direction and wind speed information and outputs wind direction and wind speed data;

[0024] S15, the image recognition sensor captures the component surface image, uses the convolutional neural network to process the collected image, distinguishes the stain type, distribution range and pollution degree, and marks its position and range in the image, and outputs the stain information data;

[0025] Furthermore, in step S3, the algorithm flow of the intelligent cleaning module is as follows:

[0026] S31, receiving instructions from the environment perception module and the adaptive control module;

[0027] S32, if the environment sensing module detects that there is dust on the surface of the component and the dust is evenly distributed, the ultrasonic vibration cleaning technology is activated to loosen and remove the dust;

[0028] S33, if the image recognition sensor finds a stain, the intelligent spray cleaning subsystem sprays an appropriate amount of cleaning agent at the stain location for cleaning, and then starts the negative pressure suction device to recover the residual material;

[0029] S34, using a flexible rotating cleaning brush to wipe and clean, to ensure that the surface of the component is clean. Further, in step S2, the algorithm flow of the adaptive control module is as follows:

[0030] S21, receiving data transmitted by the environment perception module;

[0031] S22, establishes a cleaning decision model based on an intelligent algorithm combining fuzzy logic and neural network;

[0032] S23, generating an optimal cleaning strategy in real time;

[0033] S24, sending the cleaning strategy instruction to the intelligent cleaning module.

[0034] Furthermore, in step S4, the algorithm flow of the energy recovery module is as follows:

[0035] S41, during the cleaning process, recovers the energy generated by the vibration and air flow movement on the surface of the photovoltaic module.

[0036] S42, converts mechanical vibration energy and airflow kinetic energy into electrical energy through a micro energy harvester mounted on the module frame or cleaning device;

[0037] S43, storing the electric energy in an energy storage device;

[0038] S44, when the power in the energy storage device reaches a certain threshold, the recovered energy is used first to power the low-power devices in the environmental perception module and the intelligent cleaning module; when the energy is insufficient, it switches to the output power of the external power supply or the photovoltaic module; wherein, the recovered energy is integrated with the overall energy management system of the photovoltaic power station to achieve optimal utilization and allocation of energy.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] (1) The present invention combines multi-sensor fusion environmental perception technology and intelligent cleaning technology to more accurately identify the type and degree of contamination, and adopt targeted cleaning methods. Compared with traditional cleaning technologies, the cleaning effect is significantly improved. For example, it is more effective in removing fine dust and stubborn stains, which can increase the power generation efficiency of photovoltaic modules by 8%, because a cleaner module surface can more effectively receive sunlight and convert it into electricity more efficiently.

[0041] (2) The present invention reduces the labor cost of manual cleaning and the maintenance cost of mechanical automatic cleaning devices due to frequent replacement of vulnerable parts. At the same time, the application of the energy recovery module reduces the energy consumption of the system, further saving operating costs. Assuming that the manual cleaning cost is 200 yuan per person per day and the equipment loss cost is 5,000 yuan per year, after comprehensive calculation, the cleaning and maintenance cost can be reduced by 30%.

[0042] (3) The ultrasonic vibration cleaning and flexible rotating cleaning brush in the intelligent cleaning module of the present invention avoid scratching and damaging the surface of the module, thereby extending the service life of the photovoltaic module. Compared with traditional cleaning methods that may damage the module, the cleaning system of the present invention can better maintain the performance and integrity of the module.

[0043] (4) The adaptive control algorithm based on fuzzy logic and neural network enables the system to automatically adjust the cleaning strategy according to different environments and pollution conditions, improves the intelligence level and operation efficiency of the system, and reduces the need for manual intervention. Compared with the traditional cleaning method controlled by fixed procedures or simple rules, the system of the present invention can more flexibly adapt to various complex situations and improve the timeliness and accuracy of cleaning.

[0044] (5) The energy recovery module of the present invention realizes energy recycling and improves the energy utilization efficiency and sustainability of the system. It reduces the dependence on external power supply to a certain extent, enhances the stability and reliability of the cleaning system under different environmental conditions, and is conducive to the long-term stable operation of the photovoltaic power station. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a system block diagram of the present invention.

[0046] Figure 2 Schematic diagram of the structure of the intelligent cleaning module in an embodiment of the present invention.

[0047] The names corresponding to the reference numerals are:

[0048] 1- Photovoltaic module, 2- Environmental perception module, 3- Adaptive control module, 4- Energy recovery module, 5- Ultrasonic vibration cleaning subsystem, 6- Intelligent jet cleaning subsystem, 7- Image recognition sensor, 8- Ultrasonic transducer, 9- Flexible rotating cleaning brush system, 10- Nozzle, 11- Negative pressure suction device. DETAILED DESCRIPTION

[0049] The present invention is further described below in conjunction with the accompanying drawings and embodiments. The embodiments of the present invention include but are not limited to the following embodiments.

[0050] Example

[0051] like Figure 1 As shown, the present invention discloses an intelligent adaptive cleaning system for photovoltaic components, including an environmental sensing module for detecting the cleaning status of the photovoltaic components, an intelligent cleaning module connected to the environmental sensing module for performing a cleaning function, an adaptive control connected to the environmental sensing module for controlling the intelligent cleaning module, and an energy recovery module connected to the intelligent cleaning module for realizing energy recovery and feeding back power to the environmental sensing module.

[0052] In this embodiment, the environment perception module is specifically:

[0053] Dust concentration sensor: uses the principle of laser scattering to accurately measure the concentration of dust particles in the air near the component surface. When the laser irradiates the dust particles in the air, the dust concentration is calculated based on information such as the intensity and angle of the scattered light.

[0054] Humidity sensor: monitors the ambient humidity in real time, as humidity has an important impact on the adhesion of dust and the formation of stains.

[0055] Rainfall sensor: Accurately detects whether there is rainfall and the amount of rainfall.

[0056] Wind direction and speed sensor: Obtain wind direction and speed information to optimize cleaning strategies and utilize the power of natural wind during cleaning.

[0057] Image recognition sensor: By taking images of the component surface, the sensor uses a convolutional neural network to process the collected images. After training with a large number of samples, it can accurately distinguish the type of stains (such as dust, bird droppings, leaves, etc.), distribution range and degree of contamination, and accurately mark their location and range in the image.

[0058] In this embodiment, the sensor transmits the collected data to the adaptive control module in real time through the wireless communication module.

[0059] like Figure 2 As shown, in this embodiment, the environment perception module 2, the adaptive control module 3, the energy recovery module 4, and the intelligent cleaning module are all installed on the photovoltaic module 1. The intelligent cleaning module includes an ultrasonic vibration cleaning subsystem 5, an intelligent jet cleaning subsystem 6, a flexible rotating cleaning brush system 9, an ultrasonic transducer 8, a negative pressure suction device 11, and a nozzle 10 installed on the photovoltaic module 1; wherein, the ultrasonic vibration cleaning subsystem 5 realizes ultrasonic vibration cleaning through the ultrasonic transducer 8; the image recognition sensor 7 recognizes the stains, the intelligent jet cleaning subsystem 6 cleans the stains by spraying a cleaning agent through a nozzle 10, and recovers the residual substances by using the negative pressure suction device 11, and finally wipes and cleans by using the flexible rotating cleaning brush system.

[0060] Specifically, ultrasonic vibration cleaning: a special ultrasonic transducer is installed on the surface of the component. When powered on, the transducer generates high-frequency ultrasonic vibrations, which loosen and detach the dust and stains on the surface of the component under the action of vibration. This cleaning method has a good cleaning effect on fine dust particles and stains with strong adhesion, and will not cause damage to the surface of the component.

[0061] Smart jet cleaning: For stubborn stains such as bird droppings and oil stains, the smart jet cleaning system is equipped. The system automatically locates the stains based on the detection results of the image recognition sensor and accurately controls the spray volume and pressure of the detergent. The detergent uses an environmentally friendly and efficient formula that can quickly decompose and remove stubborn stains. After jet cleaning, the negative pressure suction device is used to promptly suck and recover the remaining stains and detergent to prevent secondary pollution.

[0062] Flexible rotating cleaning brush: The bristles of the cleaning brush are made of special soft material, which can effectively remove various stains without scratching the surface of the component. The rotation speed and angle of the cleaning brush can be intelligently adjusted according to the data provided by the environmental perception module and the instructions of the adaptive control module. For example, when the wind speed is high, the rotation speed of the cleaning brush is appropriately reduced to avoid poor cleaning effect or additional pressure on the component due to the influence of wind; when the stains are unevenly distributed, the angle of the cleaning brush is adjusted to ensure comprehensive and even cleaning of the component surface.

[0063] Cleaning process example: When the environmental perception module detects that there is a lot of dust on the surface of the component and the dust is distributed relatively evenly, the intelligent cleaning module first activates the ultrasonic vibration cleaning technology to loosen the dust; if the image recognition sensor finds local stubborn stains, such as bird droppings, the intelligent jet cleaning system will spray an appropriate amount of detergent at the stain location for cleaning, and then activate the negative pressure suction device to recover the residual material, and finally use the flexible rotating cleaning brush for further wiping and cleaning to ensure that the component surface is clean.

[0064] The adaptive control module establishes a cleaning decision model based on an intelligent algorithm combining fuzzy logic and neural networks. The model comprehensively considers multiple factors such as dust concentration, humidity, rainfall, wind direction and speed, stain type and pollution degree, and generates the optimal cleaning strategy in real time.

[0065] Control strategy example: When the dust concentration is high and the humidity is low, a combination of ultrasonic vibration cleaning and intelligent jet cleaning is preferred, and the working parameters of the cleaning device are adjusted according to the wind speed to improve cleaning efficiency. When the rainfall is heavy, the control module will suspend the cleaning operation and use rainwater to naturally rinse the component surface, while adjusting the inclination angle of the component to ensure that rainwater can successfully carry away stains.

[0066] During the cleaning process, the energy generated by the vibration and airflow movement on the surface of the photovoltaic module is recovered. For example, under the action of ultrasonic vibration cleaning and natural wind, the mechanical vibration energy and airflow kinetic energy are converted into electrical energy through micro energy collectors (such as piezoelectric energy collectors and electromagnetic energy collectors) installed on the module frame or cleaning device, and stored in energy storage devices (such as supercapacitors or rechargeable batteries).

[0067] Energy application and integration: The recovered energy is used to power the sensors in the environmental perception module, the microprocessor in the adaptive control module, and some low-power devices in the intelligent cleaning module, reducing the dependence on external power supplies, and reducing the energy consumption and operating costs of the system. At the same time, the energy recovery module can also be integrated with the overall energy management system of the photovoltaic power station to achieve optimal utilization and allocation of energy. For example, when the power in the energy storage device reaches a certain threshold, the recovered energy can be used to power the cleaning system first, and then switch to the external power supply or the output power of the photovoltaic module when the energy is insufficient.

[0068] The algorithm flow of each module in this embodiment is as follows:

[0069] 1. Environmental perception module algorithm flow:

[0070] The dust concentration sensor uses the principle of laser scattering to measure the concentration of dust particles in the air near the component surface, calculates the dust concentration based on information such as the intensity and angle of the scattered light, and outputs the data.

[0071] The humidity sensor monitors the ambient humidity in real time and outputs humidity data.

[0072] The rainfall sensor accurately detects whether there is rainfall and the amount of rainfall, and outputs rainfall information data.

[0073] The wind direction and wind speed sensor obtains wind direction and wind speed information and outputs wind direction and wind speed data.

[0074] The image recognition sensor captures the component surface image and uses the convolutional neural network to process the collected image, distinguish the stain type, distribution range and degree of contamination, mark its position and range in the image, and output the stain information data.

[0075] The data collected by all the above sensors are transmitted to the adaptive control module in real time through the wireless communication module.

[0076] 2. Intelligent cleaning module algorithm process:

[0077] Receive instructions from the environment perception module and the adaptive control module.

[0078] If the environmental sensing module detects that there is a lot of dust on the surface of the component and it is evenly distributed, the ultrasonic vibration cleaning technology is activated to loosen the dust.

[0079] If the image recognition sensor finds local stubborn stains, such as bird droppings, the intelligent jet cleaning system will spray an appropriate amount of detergent at the stain location for cleaning, and then start the negative pressure suction device to recover the residual substances.

[0080] Finally, use a flexible rotating cleaning brush to further wipe and clean to ensure that the component surface is clean.

[0081] The rotation speed and angle of the cleaning brush are intelligently adjusted based on the data provided by the environmental perception module and the instructions of the adaptive control module, such as reducing the rotation speed when the wind speed is high and adjusting the angle when the stains are unevenly distributed.

[0082] 3. Adaptive control module algorithm flow:

[0083] Receive data transmitted by the environmental sensing module, including information such as dust concentration, humidity, rainfall, wind direction and speed, stain type and pollution degree.

[0084] Establish a cleaning decision model based on an intelligent algorithm combining fuzzy logic and neural networks

[0085] Comprehensively consider multiple factors and generate the optimal cleaning strategy in real time. When the dust concentration is high and the humidity is low, the combination of ultrasonic vibration cleaning and intelligent jet cleaning is preferred, and the working parameters of the cleaning device are adjusted according to the wind speed; when the rainfall is heavy, the cleaning operation is suspended, and the tilt angle of the components is adjusted to use rainwater for natural washing.

[0086] Send cleaning strategy instructions to the smart cleaning module.

[0087] 4. Energy recovery module algorithm flow:

[0088] During the cleaning process, the energy generated by the vibration of the photovoltaic panel surface and the movement of air flow is used.

[0089] Convert mechanical vibration energy and airflow kinetic energy into electrical energy through micro energy harvesters (such as piezoelectric energy harvesters and electromagnetic energy harvesters) installed on the component frame or cleaning device.

[0090] Storing electrical energy in an energy storage device (such as a supercapacitor or rechargeable battery).

[0091] When the amount of electricity in the energy storage device reaches a certain threshold, the recovered energy is used first to power the sensors in the environmental perception module, the microprocessor in the adaptive control module, and some low-power devices in the intelligent cleaning module; when the energy is insufficient, it switches to the output power of the external power supply or the photovoltaic module. At the same time, it is integrated with the overall energy management system of the photovoltaic power station to achieve optimal utilization and allocation of energy.

[0092] During specific implementation, the parameter configuration of each module is as follows:

[0093] 1. Intelligent cleaning module

[0094] Ultrasonic vibration cleaning subsystem: including ultrasonic transducer (frequency 20kHz-100kHz), drive circuit (power supply voltage 50V-220V), working current 3A-10A.

[0095] Intelligent jet cleaning subsystem: with image recognition sensor (resolution 1920×1080 pixels or higher, frame rate 15fps-30fps), nozzle (diameter 0.5mm-2mm), jet pressure (0.5MPa-2MPa), detergent jet volume (0.1ml-0.5ml per square centimeter of stain), negative pressure suction device (suction pressure -0.05MPa--0.1MPa, suction flow 10L / min-50L / min).

[0096] Flexible rotating cleaning brush system: including cleaning brush (bristle length 1cm-3cm, bristle hardness Shore hardness 50A-80A), motor drive system (rotation speed 50rpm-500rpm), angle adjustment accuracy (±5°-±10°).

[0097] 2. Adaptive control module

[0098] Data collection and preprocessing: data collection frequency (environmental perception data every 5-15 minutes, and every 3-10 minutes in areas with rapid environmental changes), abnormal data threshold (abnormal conditions such as dust concentration, humidity, rainfall, etc.), noise interference removal (using filtering algorithm, filtering window 3-7 data points).

[0099] Fuzzy logic processing: fuzzy set partitioning, dust concentration: low (0-50μg / m 3 ), medium (50 μg / m 3 -150 μg / m 3 ), high (150 μg / m 3 Humidity: dry (0-30%RH), moderate (30%RH-70%RH), humid (70%RH or more). Wind direction and speed: breeze (wind speed 0-3m / s), moderate wind (wind speed 3m / s-8m / s), strong wind (wind speed 8m / s or more). Number of fuzzy rules: generally set 50-100 fuzzy rules to fully cover various possible combinations of environments and stains.

[0100] Neural network calculation: neural network structure (5-10 neurons in the input layer, 10-40 neurons in the middle layer, 3-6 neurons in the output layer), training parameters (learning rate 0.01-0.1, number of training rounds 500-2000 rounds, training data ratio 70%-80% for training, 20%-30% for verification and testing).

[0101] 3. Energy recovery module

[0102] Energy harvesting submodule:

[0103] Piezoelectric energy harvester: piezoelectric material (lead zirconate titanate piezoelectric coefficient d33 is 100pC / N-700pC / N, operating frequency is 10Hz-100Hz).

[0104] Electromagnetic energy collector: induction coil (50-200 turns), magnetic core material (magnetic permeability 500-3000), energy conversion circuit (rectification efficiency 80%-95% or more), filter capacitor (100μF-500μF).

[0105] Energy storage submodule:

[0106] Rechargeable battery: battery capacity (500mAh-5000Ah), charging cut-off voltage (lithium-ion battery 4.2V-4.3V), discharging cut-off voltage (2.5V-3.0V).

[0107] Energy allocation algorithm:

[0108] Power threshold: high power threshold (energy storage device capacity 80%-90%), low power threshold (energy storage device capacity 20%-30%).

[0109] Energy allocation priority: high-priority equipment (critical microprocessors in the adaptive control module, allocated 30%-50% of the total energy), medium-priority equipment (important sensors in the environmental perception module, allocated 20%-30%), and low-priority equipment (non-critical equipment in the intelligent cleaning module, allocated 20%-30% when energy is sufficient).

[0110] Through the above design, the present invention can more accurately identify the type and degree of stains and adopt targeted cleaning methods through the combination of multi-sensor fusion environmental perception technology and intelligent cleaning technology, and significantly improve the cleaning effect compared with traditional cleaning technology. Therefore, compared with the prior art, the present invention has outstanding substantive features and significant progress.

[0111] The above embodiment is only one of the preferred implementation modes of the present invention and should not be used to limit the protection scope of the present invention. Any changes or modifications that are made to the main design concept and spirit of the present invention and have no substantive significance, and the technical problems they solve are still consistent with the present invention, should be included in the protection scope of the present invention.

Claims

1. A photovoltaic module intelligent adaptive cleaning system, characterized in that: It includes an environmental perception module for detecting the cleaning status of photovoltaic components, an intelligent cleaning module connected to the environmental perception module for performing a cleaning function, an adaptive control connected to the environmental perception module for controlling the intelligent cleaning module, and an energy recovery module connected to the intelligent cleaning module for realizing energy recovery and feeding back power to the environmental perception module.

2. The photovoltaic module intelligent adaptive cleaning system according to claim 1, characterized in that: The intelligent cleaning module includes an ultrasonic vibration cleaning subsystem, an intelligent jet cleaning subsystem, a flexible rotating cleaning brush system, an ultrasonic transducer, a negative pressure suction device and a nozzle installed on the photovoltaic module; wherein the ultrasonic vibration cleaning subsystem realizes ultrasonic vibration cleaning through an ultrasonic transducer; the image recognition sensor identifies stains, the intelligent jet cleaning subsystem cleans the stains by spraying detergent through a nozzle, and recovers residual substances using a negative pressure suction device, and finally wipes and cleans using a flexible rotating cleaning brush system.

3. The photovoltaic module intelligent adaptive cleaning system according to claim 2, characterized in that: The environment perception module includes Dust concentration sensor: uses the principle of laser scattering to accurately measure the concentration of dust particles in the air near the component surface; Humidity sensor: real-time monitoring of ambient humidity; Rainfall sensor: accurately detects whether there is rainfall and the amount of rainfall; Wind direction and speed sensor: obtain wind direction and wind speed information; Image recognition sensor: It captures component surface images and processes them using a convolutional neural network.

4. A photovoltaic module intelligent adaptive cleaning method, characterized in that: Using the cleaning system as claimed in claim 3, S1, using the environment sensing module to detect the cleanliness status of the photovoltaic module; S2, based on the detection results of the environmental perception module, the adaptive control module uses an intelligent algorithm based on fuzzy logic and neural network to establish a cleaning decision model and generate the optimal cleaning strategy in real time; S3, according to the cleaning strategy of the adaptive control module, start the intelligent cleaning module to clean the photovoltaic module; S4, recovering the energy generated during the cleaning process through an energy recovery module.

5. A photovoltaic module intelligent adaptive cleaning method according to claim 4, characterized in that: In step S1, the working process of the environment perception module is as follows: S11, the dust concentration sensor uses the principle of laser scattering to measure the concentration of dust particles in the air near the component surface, calculates the dust concentration based on information such as the intensity and angle of the scattered light, and outputs the data; S12, the humidity sensor monitors the ambient humidity in real time and outputs humidity data; S13, the rainfall sensor accurately detects whether there is rainfall and the amount of rainfall, and outputs rainfall information data; S14, a wind direction and wind speed sensor obtains wind direction and wind speed information and outputs wind direction and wind speed data; S15, the image recognition sensor captures the component surface image, uses the convolutional neural network to process the collected image, distinguishes the stain type, distribution range and degree of contamination, and marks its position and range in the image, and outputs the stain information data.

6. A photovoltaic module intelligent adaptive cleaning method according to claim 5, characterized in that: In step S3, the algorithm flow of the intelligent cleaning module is as follows: S31, receiving instructions from the environment perception module and the adaptive control module; S32, if the environment sensing module detects that there is dust on the surface of the component and the dust is evenly distributed, the ultrasonic vibration cleaning technology is activated to loosen and remove the dust; S33, if the image recognition sensor finds a stain, the intelligent spray cleaning subsystem sprays an appropriate amount of cleaning agent at the stain location for cleaning, and then starts the negative pressure suction device to recover the residual material; S34, use a flexible rotating cleaning brush to wipe and clean to ensure that the component surface is clean.

7. A photovoltaic module intelligent adaptive cleaning method according to claim 6, characterized in that: In step S2, the algorithm flow of the adaptive control module is as follows: S21, receiving data transmitted by the environment perception module; S22, establishes a cleaning decision model based on an intelligent algorithm combining fuzzy logic and neural network; S23, generating an optimal cleaning strategy in real time; S24, sending the cleaning strategy instruction to the intelligent cleaning module.

8. The photovoltaic module intelligent adaptive cleaning method according to claim 7, characterized in that: In step S4, the algorithm flow of the energy recovery module is as follows: S41, during the cleaning process, recovers the energy generated by the vibration and air flow movement on the surface of the photovoltaic module; S42, converts mechanical vibration energy and airflow kinetic energy into electrical energy through a micro energy harvester mounted on the module frame or cleaning device; S43, storing the electric energy in an energy storage device; S44, when the power in the energy storage device reaches a certain threshold, the recovered energy is used first to power the low-power devices in the environmental perception module and the intelligent cleaning module; when the energy is insufficient, it switches to the output power of the external power supply or the photovoltaic module; wherein, the recovered energy is integrated with the overall energy management system of the photovoltaic power station to achieve optimal utilization and allocation of energy.

Images