Photovoltaic cleaning unmanned aerial vehicle, cleaning control method thereof and photovoltaic cleaning system

Abstract

The invention discloses a photovoltaic cleaning unmanned aerial vehicle, a cleaning control method thereof and a photovoltaic cleaning system, and belongs to the technical field of photovoltaic cleaning. The photovoltaic cleaning unmanned aerial vehicle comprises a cleaning system, the cleaning system comprises a first water tank, a second water tank and a spraying structure, the first water tank stores dissolving liquid, the second water tank stores cleaning liquid, and the first water tank and the second water tank communicate with the spraying structure; the method comprises the steps that for any dirty area on a photovoltaic array, a photovoltaic cleaning unmanned aerial vehicle is controlled to fly to the position above the dirty area, and a spraying structure is controlled to spray a dissolving solution to the dirty area; wherein the spraying parameters of the spraying structure for spraying the dissolving solution are determined according to the area and the pollution degree of the dirty region; when the dirty area meets the dissolving condition, the spraying structure is controlled to conduct cleaning fluid pressurization washing on the dirty area; wherein the flushing parameters of the spraying structure for cleaning fluid pressurized flushing are adjusted according to the area and the pollution degree of the dirty area. According to the embodiment of the invention, stains on the surface of the photovoltaic array can be accurately removed.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic cleaning technology, and in particular to a photovoltaic cleaning drone and its cleaning control method and photovoltaic cleaning system. Background Technology

[0002] The power generation efficiency of a photovoltaic power station is closely related to the cleanliness of the photovoltaic panels. Pollutants such as bird droppings and dust can seriously affect the power generation efficiency of photovoltaic panels. Among them, dried stains such as bird droppings and mud contain sticky components and are difficult to completely remove by conventional airflow cleaning or large-area water flow cleaning after drying. Summary of the Invention

[0003] This invention provides a photovoltaic cleaning drone and its cleaning control method and photovoltaic cleaning system to solve one of the technical problems mentioned in the background art.

[0004] In a first aspect, embodiments of the present invention provide a cleaning control method for a photovoltaic cleaning drone. The photovoltaic cleaning drone includes a cleaning system, which includes a first water tank, a second water tank, and a spraying structure. The first water tank stores a dissolving solution, the second water tank stores a cleaning solution, and the first water tank and the second water tank are respectively connected to the spraying structure.

[0005] The method includes: targeting any stained area on the photovoltaic array:

[0006] The photovoltaic cleaning drone is controlled to fly above the stained area, and the spraying structure is controlled to spray a dissolving liquid onto the stained area; wherein, during the dissolving liquid spraying process, the spraying parameters of the spraying structure for spraying the dissolving liquid are determined according to the area and degree of contamination of the stained area.

[0007] When the stained area meets the dissolution conditions, the spraying structure is controlled to pressurize and rinse the stained area with cleaning fluid; wherein, during the rinsing process, the rinsing parameters of the spraying structure for pressurizing and rinsing with cleaning fluid are adjusted according to the area and degree of contamination of the stained area.

[0008] Optionally, the spraying structure includes: a diaphragm pump, a nozzle, and a nozzle angle adjustment mechanism; the first water tank is connected to the nozzle through the diaphragm pump;

[0009] The spraying parameters for the dissolving solution are determined based on the area and degree of contamination of the stained area, including:

[0010] The flow rate of the diaphragm pump is adjusted according to the area of ​​the stained area;

[0011] Based on the relative position of the nozzle and the stained area, the nozzle angle adjustment mechanism is controlled to adjust the angle of the nozzle so that the dissolving liquid sprayed by the nozzle covers the stained area;

[0012] The spraying time of the dissolving solution from the nozzle is adjusted according to the degree of contamination of the stained area; wherein, the higher the degree of contamination, the longer the spraying time.

[0013] Optionally, the spraying structure includes: a pressure pump and a nozzle; the second water tank is connected to the nozzle through the pressure pump;

[0014] The rinsing parameters for pressurized rinsing with cleaning fluid are adjusted according to the area and degree of contamination of the stained area, including:

[0015] Adjust the nozzle diameter according to the area of ​​the stained area;

[0016] Adjust the flushing water pressure applied to the nozzle by the pressurization pump and / or adjust the flushing time of the nozzle according to the degree of contamination of the stained area.

[0017] Optionally, the photovoltaic cleaning drone further includes: a visual recognition system;

[0018] Adjusting the flushing water pressure supplied to the nozzle by the pressurized pump and / or adjusting the flushing time of the nozzle according to the degree of contamination in the stained area includes:

[0019] The lower limit flushing water pressure, the upper limit flushing water pressure, and the maximum flushing time are determined based on the degree of contamination in the stained area.

[0020] The pressurization pump is controlled to provide the lower limit water pressure and the nozzle is opened to spray cleaning fluid onto the stained area, thus initiating the rinsing of the stained area;

[0021] During the rinsing process, at each first preset time interval, the visual recognition system is controlled to acquire an image of the stained area and compare it with the image of the stained area before rinsing. The actual stain removal effect is obtained based on a preset cleanliness evaluation model. If the actual stain removal effect does not reach the preset stain removal effect, rinsing continues at the current rinsing water pressure or the rinsing water pressure is increased to continue rinsing until the actual stain removal effect reaches the preset stain removal effect or the rinsing time reaches the maximum rinsing time, at which point rinsing of the stained area is stopped. The rinsing water pressure used during the rinsing process does not exceed the upper limit rinsing water pressure.

[0022] Optionally, the spraying structure further includes: a pressure sensor for detecting the water pressure at the nozzle; the method further includes:

[0023] During the rinsing process, the rinsing water pressure provided by the pressurizing pump is adjusted based on the actual water pressure feedback of the nozzle detected by the pressure sensor.

[0024] Optionally, the spraying structure further includes: a three-way solenoid valve, wherein the first water tank and the second water tank are respectively connected to the nozzle through the three-way solenoid valve; the method further includes:

[0025] When it is necessary to spray the dissolving solution, control the three-way solenoid valve to connect the first water tank and the nozzle;

[0026] When pressurized rinsing with cleaning fluid is required, the three-way solenoid valve is controlled to connect the second water tank and the nozzle.

[0027] Optionally, the cleaning system further includes: a first liquid level sensor and a second liquid level sensor, respectively used to detect the liquid level in the first water tank and the liquid level in the second water tank; the method further includes:

[0028] When the first liquid level sensor detects that the actual liquid level in the first water tank is lower than the first preset liquid level, the photovoltaic cleaning drone is controlled to return to the take-off and landing point to add dissolving liquid to the first water tank.

[0029] When the second liquid level sensor detects that the actual liquid level in the second water tank is lower than the second preset liquid level, the photovoltaic cleaning drone is controlled to return to the take-off and landing point to add cleaning fluid to the second water tank.

[0030] Optionally, the surface of the photovoltaic array includes a plurality of the stained areas; the method includes:

[0031] Obtain a stain cleaning list; wherein, the stain cleaning list includes the cleaning order of each stain area on the photovoltaic array;

[0032] According to the stain cleaning list, the photovoltaic cleaning drone is controlled to fly sequentially above each of the stain areas, and the spraying structure is controlled to spray dissolving liquid onto each of the stain areas;

[0033] According to the stain cleaning list, the photovoltaic cleaning drone is controlled to fly sequentially above each of the stain areas, and when the stain area meets the dissolution conditions, the spraying structure is controlled to pressurize and rinse the stain area with cleaning fluid.

[0034] Optionally, the dissolution conditions include: the dissolution time after spraying the dissolving solution reaches a second preset time, and / or the dissolution effect reaches a preset effect.

[0035] Optionally, the photovoltaic cleaning drone also includes a visual recognition module;

[0036] The method further includes: for any of the stained areas:

[0037] After spraying the dissolving solution onto the stained area and controlling the photovoltaic cleaning drone to fly over the stained area again, the visual recognition module is controlled to collect an image of the stained area every third preset time interval, and the dissolving effect of the stained area is determined by the preset dissolving effect judgment model; wherein, the third preset time interval is less than the second preset time interval.

[0038] Optionally, obtaining the stain cleaning list includes:

[0039] Obtain the location of each of the stain areas on the photovoltaic array;

[0040] Select any of the aforementioned stained areas as the starting stained area, and sort the stained areas according to the shortest path principle to form the stained cleaning list.

[0041] Optionally, the photovoltaic cleaning drone includes: a visual recognition system and a positioning system;

[0042] Obtaining the location of each of the stain areas on the photovoltaic array includes:

[0043] The system controls the photovoltaic cleaning drone to fly along a preset route and controls the visual recognition system to cooperate with the positioning system to identify each stain area and its location on the surface of the photovoltaic array.

[0044] Specifically, the first stained area on the preset route is selected as the starting stained area.

[0045] Optionally, after the dissolving solution has been sprayed on all the stained areas, the method further includes:

[0046] Determine whether the battery life of the photovoltaic cleaning drone meets the preset battery life conditions;

[0047] If so, control the photovoltaic cleaning drone to fly above the starting stain area in the stain cleaning list;

[0048] If not, the photovoltaic cleaning drone is controlled to return to the take-off and landing point for charging, and after charging is completed, it flies to the area above the starting stain in the stain cleaning list.

[0049] Optionally, after pressurizing and rinsing all the stained areas with cleaning fluid, the method further includes:

[0050] In the case of residual stains on the surface of the photovoltaic array, the steps of spraying dissolving liquid and pressurizing rinsing with cleaning liquid are performed on each residual stain area; wherein, the residual stain area is the stain area where the difference between the image grayscale after pressurizing rinsing with cleaning liquid and the image grayscale of the clean area is greater than a preset threshold.

[0051] If there are no residual stains on the surface of the photovoltaic array, control the photovoltaic cleaning drone to return to the take-off and landing point to complete the cleaning operation.

[0052] Secondly, embodiments of the present invention also provide a photovoltaic cleaning drone, comprising: a cleaning system and a control unit; the cleaning system comprising: a first water tank, a second water tank and a spraying structure, wherein the first water tank stores a dissolving solution, the second water tank stores a cleaning solution, and the first water tank and the second water tank are respectively connected to the spraying structure; the control unit is used to execute the cleaning control method of the photovoltaic cleaning drone provided in any embodiment of the present invention.

[0053] Optionally, the photovoltaic cleaning drone further includes: a wireless communication module connected to the control unit; wherein the control unit communicates with the ground station through the wireless communication module.

[0054] Thirdly, embodiments of the present invention also provide a photovoltaic cleaning system, including: a ground station and a photovoltaic cleaning drone provided in any embodiment of the present invention.

[0055] The cleaning control method for photovoltaic cleaning drones provided in this invention is based on a photovoltaic cleaning drone equipped with dual water tanks. It can perform targeted cleaning of soiled areas on the surface of photovoltaic arrays, and is particularly suitable for the precise removal of stubborn stains that are difficult to remove using large-area airflow or water flow. Specifically, for any soiled area, a dissolving solution is first sprayed to dissolve the stain. Once the dissolution conditions are met, pressurized rinsing with cleaning solution is performed. Both the spraying and rinsing parameters are adjusted according to the actual characteristics of the soiled area (including area and degree of contamination). This provides an automated stain cleaning solution that can accurately identify, efficiently dissolve, and adaptively rinse. Pressurized rinsing after stain dissolution ensures stain removal effectiveness by applying the dissolving solution and avoids stain residue after dissolution by pressurized rinsing. Furthermore, this embodiment, based on a photovoltaic cleaning drone for targeted and automated cleaning of each soiled area, effectively improves cleaning efficiency and avoids the personal safety hazards associated with manual cleaning. It also eliminates the need for high-pressure water spraying over the entire area, thus avoiding water waste.

[0056] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of the structure of the photovoltaic cleaning drone provided in an embodiment of the present invention;

[0059] Figure 2 This is a flowchart of the cleaning control method for a photovoltaic cleaning drone provided in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the cleaning system provided in an embodiment of the present invention;

[0061] Figure 4 This is a pressure-flow characteristic curve of the nozzle provided in an embodiment of the present invention;

[0062] Figure 5 This is a schematic diagram of the cleaning operation path planning provided in an embodiment of the present invention;

[0063] Figure 6 This is another flowchart of the cleaning control method for the photovoltaic cleaning drone provided in this embodiment of the invention;

[0064] Figure 7 This is another flowchart of the cleaning control method for the photovoltaic cleaning drone provided in the embodiments of the present invention;

[0065] Figure 8 This is another flowchart of the cleaning control method for the photovoltaic cleaning drone provided in the embodiments of the present invention;

[0066] Figure 9 These are comparison images of the cleaning effects of the photovoltaic cleaning drone provided in this embodiment of the invention;

[0067] Figure 10 This is a schematic diagram of the structure of the photovoltaic cleaning system provided in an embodiment of the present invention. Detailed Implementation

[0068] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0069] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0070] This invention provides a cleaning control method for a photovoltaic cleaning drone, which can achieve precise cleaning of stains on the surface of a photovoltaic array. For ease of explanation, the structure of the photovoltaic cleaning drone will be briefly described below.

[0071] Figure 1 This is a schematic diagram of the structure of the photovoltaic cleaning drone provided in an embodiment of the present invention. See also... Figure 1 The photovoltaic cleaning drone may include a flight control system consisting of a flight platform 500, which is composed of multiple rotors, such as quadcopters or hexacopters, used to drive the entire photovoltaic cleaning drone in flight and hovering.

[0072] Furthermore, the photovoltaic cleaning drone also includes a cleaning system. The cleaning system includes a first water tank 110, a second water tank 120, and a spraying structure 130; wherein the first water tank 110 stores a dissolving solution, which may contain surfactants and biological enzymes; the second water tank 120 stores a cleaning solution, which may be, for example, clean water, specifically filtered clean water, used to rinse dissolved stains; the spraying structure 130 sprays liquid outward through nozzles; the first water tank 110 and the second water tank 120 are respectively connected to the spraying structure 130, for example... Figure 1 As shown by the dashed line, the first water tank 110 and the second water tank 120 are respectively connected to the spray structure 130 through independent liquid supply pipelines.

[0073] Furthermore, the photovoltaic cleaning drone also includes a control unit 200, which serves as the control core of the photovoltaic cleaning drone and can be used to control other functional modules, such as controlling the flight control system and executing the cleaning control method provided in the embodiments of the present invention. Additionally, the photovoltaic cleaning drone may also include a visual recognition system 300 and a positioning system 400, both of which can be controlled by the control unit 200. The visual recognition system 300 can be used to identify stain areas, and the positioning system 400 can be used to locate the drone's position and the location of the stain areas.

[0074] The cleaning control method for photovoltaic cleaning drones is described below. This method is applicable to the precise removal of dried, stubborn stains such as bird droppings and mud from the surface of photovoltaic arrays, and can be executed by the control unit configured in the photovoltaic cleaning drone. Figure 2 This is a flowchart of the cleaning control method for a photovoltaic cleaning drone provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the cleaning control method of the photovoltaic cleaning drone includes: performing the following steps on any soiled area on the photovoltaic array:

[0075] S110: Control the photovoltaic cleaning drone to fly above the stained area and control the spraying structure to spray the dissolving liquid onto the stained area.

[0076] During the dissolving solution spraying process, the spraying parameters are determined based on the area and degree of contamination of the stained area. Specific spraying parameters may include at least one of the following: flow rate, time, area, angle, and pressure of the sprayed dissolving solution, which can be adjusted as needed. For example, controlling the photovoltaic cleaning drone to fly above the stained area can mean controlling the drone to fly directly above the stained area to ensure precise spraying of the dissolving solution.

[0077] The stained areas can specifically refer to areas on the photovoltaic array surface covered with stubborn stains. Stubborn stains can be understood as stains that are difficult to remove using conventional airflow cleaning or large-area water flow cleaning, such as dried stains like bird droppings or mud. Information on the stained areas on the photovoltaic array surface (such as the number of stained areas, the location, area, and degree of contamination of each stained area) can be collected by other inspection equipment and provided to the photovoltaic cleaning drone, or it can be collected by the photovoltaic cleaning drone itself; no specific limitation is made here.

[0078] S120. When the stain area meets the dissolution conditions, control the spray structure to pressurize and rinse the stain area with cleaning fluid.

[0079] During the rinsing process, the rinsing parameters are adjusted based on the area and degree of contamination of the stained area to pressurize and rinse the cleaning fluid. Specific rinsing parameters may include at least one of the following: rinsing pressure, rinsing time, cleaning fluid flow rate, cleaning fluid spray angle, and cleaning fluid spray area. For example, when rinsing a stained area, the photovoltaic cleaning drone can be controlled to fly above the stained area, specifically directly above it, to ensure precise rinsing. For example, the dissolution conditions may include at least one of meeting the dissolution time or meeting the dissolution effect standards, which can be set according to actual needs.

[0080] The cleaning control method for photovoltaic cleaning drones provided in this invention is based on a photovoltaic cleaning drone equipped with dual water tanks. It can perform targeted cleaning of soiled areas on the surface of photovoltaic arrays, and is particularly suitable for the precise removal of stubborn stains that are difficult to remove using large-area airflow or water flow. Specifically, for any soiled area, a dissolving solution is first sprayed to dissolve the stain. Once the dissolution conditions are met, pressurized rinsing with cleaning solution is performed. Both the spraying and rinsing parameters are adjusted according to the actual characteristics of the soiled area (including area and degree of contamination). This provides an automated stain cleaning solution that can accurately identify, efficiently dissolve, and adaptively rinse. Pressurized rinsing after stain dissolution ensures stain removal effectiveness by applying the dissolving solution and avoids stain residue after dissolution by pressurized rinsing. Furthermore, this embodiment, based on a photovoltaic cleaning drone for targeted and automated cleaning of each soiled area, effectively improves cleaning efficiency and avoids the personal safety hazards associated with manual cleaning. It also eliminates the need for high-pressure water spraying over the entire area, thus avoiding water waste.

[0081] The following description, in conjunction with the specific structure of the cleaning system, provides an exemplary illustration of the implementation of each step of this method, but this is not intended to limit the scope of the invention.

[0082] Figure 3 This is a schematic diagram of the cleaning system provided in an embodiment of the present invention. See also... Figure 3 In one embodiment, the spraying structure optionally includes a diaphragm pump 132, a nozzle 131, and a nozzle angle adjustment mechanism 133. The first water tank 110 is connected to the nozzle 131 via the diaphragm pump 132. For example, the first water tank is connected to the nozzle 131 via a first liquid supply line. The diaphragm pump 132 is located on the first liquid supply line. The control unit 200 can control the flow rate of the dissolving solution sprayed by the nozzle 131 by controlling the flow rate of the diaphragm pump 132. Specifically, the diaphragm pump 132 can be a miniature diaphragm pump.

[0083] Specifically, the spraying parameters for the dissolving solution are determined based on the area and degree of contamination of the stained area, including:

[0084] 1) Adjust the flow rate of the diaphragm pump according to the area of ​​the stained area.

[0085] For example, the larger the area of ​​the stain, the greater the flow rate of the diaphragm pump 132, so as to achieve full coverage and thorough wetting of the stain by the dissolving liquid.

[0086] 2) Adjust the nozzle angle according to the relative position of the nozzle and the stained area, so that the dissolving liquid sprayed by the nozzle covers the stained area.

[0087] For example, the nozzle angle adjustment mechanism 133 may include a servo motor for adjusting the nozzle angle. Specifically, the nozzle 131 can atomize the dissolving liquid and spray it onto the stained area. When the nozzle 131 is directly above the centroid of the stained area, the nozzle angle can be controlled to point directly downwards; when the nozzle 131 is deviated from directly above the centroid of the stained area, the nozzle angle can be controlled to point towards the centroid of the stained area. This allows the atomized dissolving liquid to precisely cover the stained area.

[0088] 3) Adjust the spraying time of the dissolving solution according to the degree of contamination of the stained area.

[0089] The higher the degree of contamination, the longer the spraying time. By setting the spraying time according to the degree of contamination, it can be ensured that the dissolving solution fully penetrates the stains.

[0090] In summary, it enables precise, targeted spraying of the dissolving solution onto stained areas.

[0091] See also Figure 3 In one embodiment, optionally, the spraying structure includes: a pressure pump 134 and a nozzle 131; a second water tank 120 is connected to the nozzle 131 via the pressure pump 134. For example, the second water tank 120 is connected to the nozzle 131 via a second liquid supply line, the pressure pump 134 is disposed on the second liquid supply line, and the control unit 200 can control the rinsing pressure of the cleaning fluid sprayed by the nozzle 131 by controlling the pressure of the pressure pump 134; the pressure pump 134 can specifically be a high-pressure pump, and its pressure adjustment range can be set according to actual needs.

[0092] Specifically, the rinsing parameters for pressurized rinsing with cleaning fluid are adjusted according to the area and degree of contamination of the stained area, including:

[0093] 1) Adjust the nozzle diameter according to the area of ​​the stained area.

[0094] For example, the larger the area of ​​the stain, the larger the diameter of the nozzle 131 should be to ensure that the cleaning fluid flow fully covers the stain area.

[0095] 2) Adjust the flushing water pressure applied to the nozzle by the booster pump and / or adjust the flushing time of the nozzle according to the degree of contamination in the stained area.

[0096] For example, the higher the degree of contamination, the higher the rinsing water pressure and / or the longer the rinsing time should be, in order to effectively rinse the dissolved stains and minimize stain residue.

[0097] Furthermore, photovoltaic cleaning drones can also be used as... Figure 1 As shown, it includes a visual recognition system 300. The visual recognition system 300 can determine the actual stain removal effect of the stained area by acquiring images of the stained area and based on a pre-trained preset cleanliness assessment model.

[0098] Accordingly, the flushing water pressure supplied to the nozzle by the pressurized pump and / or the flushing time of the nozzle can be adjusted according to the degree of contamination in the stained area. Specifically, this may include:

[0099] 1) Determine the lower limit of rinsing water pressure, the upper limit of rinsing water pressure, and the maximum rinsing time based on the degree of contamination of the stained area.

[0100] For example, the higher the degree of contamination, the higher the corresponding lower limit flushing water pressure and upper limit flushing water pressure, and the longer the maximum flushing time, in order to ensure the removal effect of stains.

[0101] 2) Control the pressurization pump to provide the lower limit water pressure and open the nozzle to spray cleaning fluid onto the stained area, starting the flushing of the stained area.

[0102] 3) During the rinsing process, at each first preset time interval, the visual recognition system is controlled to collect images of the stained area and compare them with images of the stained area before rinsing. The actual stain removal effect is obtained based on the preset cleanliness evaluation model. If the actual stain removal effect does not reach the preset stain removal effect, rinsing continues at the current rinsing water pressure or the rinsing water pressure is increased to continue cleaning until the actual stain removal effect reaches the preset stain removal effect or the rinsing time reaches the maximum rinsing time, at which point rinsing of the stained area is stopped. The rinsing water pressure used during the rinsing process shall not exceed the upper limit rinsing water pressure.

[0103] The first preset time can be set according to actual needs. For example, the maximum rinsing time can be set to at least five times the first preset time to provide sufficient water pressure / duration adjustment margin. The preset cleanliness assessment model is a pre-trained cleanliness assessment model, which can be optimized and trained based on the YOLO algorithm. The preset cleanliness assessment model can use a "difference comparison algorithm" to compare the difference in the degree of contamination of the stained area before and after rinsing to determine the actual stain removal effect. The preset stain removal effect is the expected reduction in the degree of contamination of the stained area after rinsing compared to the stained area before rinsing, expressed as a percentage. If the degree of contamination of the stained area after rinsing decreases by a preset percentage (e.g., at least 50%) compared to before rinsing, it indicates that the actual stain removal effect has reached the preset stain removal effect. For example, the preset stain removal effect can also be different for different degrees of contamination.

[0104] In this embodiment, the rinsing of the dissolved stain area is initiated with a limited rinsing water pressure. If the actual stain removal effect is not satisfactory, the rinsing water pressure and / or rinsing time are gradually increased. The rinsing water pressure for the stained area does not exceed the upper limit, and the rinsing time does not exceed the maximum rinsing time. This effectively balances water consumption and cleaning time, completing the rinsing with the lowest possible rinsing water pressure and shortest possible rinsing time, thereby minimizing water waste. Furthermore, by comparing images of the stained area before and after rinsing, the effect of a single rinse can be effectively determined. Compared to directly comparing images of a clean area, this provides a more flexible way to determine whether the stained area has been effectively cleaned. Especially for heavily soiled areas, comparing images of a clean area may not show complete removal, but comparing images before and after rinsing confirms effective cleaning as long as the preset stain removal effect is achieved, allowing the rinsing of the next stained area to proceed, avoiding prolonged rinsing of the same stained area.

[0105] It should be noted that, Figure 3 The example shows a first water tank 110 and a second water tank 120 connected to the same nozzle 131, but this is not intended to limit the invention. In other embodiments, the first water tank 110 and the second water tank 120 may be connected to different nozzles 131 to respectively achieve the spraying of the dissolving liquid and the cleaning liquid.

[0106] Furthermore, the capacity of the two water tanks can be set according to actual needs. Figure 4 This is a pressure-flow characteristic curve of the nozzle provided in an embodiment of the present invention. Figure 4The two shaded areas indicate the pressure and flow ranges that may be used in the dissolving solution spraying process and the pressure flushing process with the cleaning solution, respectively. For example, the dissolving solution spraying mainly uses a pressure range of 0.1-0.5 MPa, while the pressure flushing with the cleaning solution mainly uses a pressure range of 5-7 MPa. Therefore, it can be seen that pressure flushing requires a larger water volume; thus, the volume of the second water tank 120 can be set larger than that of the first water tank 110, for example, the first water tank 110 is a 2L tank, and the second water tank 120 is an 8L tank.

[0107] See also Figure 3 Based on the above embodiments, optionally, the spraying structure further includes a pressure sensor 135, which detects the water pressure of the nozzle 131; and the pressure sensor 135 is connected to the control unit 200. Correspondingly, the method further includes adjusting the rinsing water pressure provided by the booster pump based on the actual water pressure of the nozzle detected by the pressure sensor during the rinsing process. In this way, by adjusting the rinsing water pressure, the actual rinsing water pressure provided to the soiled area can better match the expected rinsing water pressure, improving the rinsing effect.

[0108] Based on the above embodiments, optionally, the first water tank 110 and the second water tank 120 are connected to the same nozzle 131, and the spraying structure further includes a three-way solenoid valve 136. The first water tank 110 and the second water tank 120 are respectively connected to the nozzle 131 through the three-way solenoid valve 136, and the three-way solenoid valve 136 is connected to the control unit 200. Accordingly, the method further includes:

[0109] 1) When it is necessary to spray the dissolving solution, control the three-way solenoid valve 136 to connect the first water tank 110 and the nozzle 131, for example, to connect the liquid supply pipeline between the first water tank 110 and the nozzle 131, so that the dissolving solution can be smoothly delivered to the nozzle 131.

[0110] 2) When pressurized rinsing with cleaning fluid is required, control the three-way solenoid valve 136 to connect the second water tank 120 and the nozzle 131, for example, to connect the liquid supply pipeline between the second water tank 120 and the nozzle 131, so that the cleaning fluid can be smoothly delivered to the nozzle 131.

[0111] Based on the above embodiments, optionally, the cleaning system further includes: a first liquid level sensor 111 and a second liquid level sensor 121. The first liquid level sensor 111 is, for example, disposed in the first water tank 110, for detecting the liquid level in the first water tank 110; the second liquid level sensor 121 is, for example, disposed in the second water tank 120, for detecting the liquid level in the second water tank 120. Correspondingly, the method further includes:

[0112] 1) When the first liquid level sensor detects that the actual liquid level in the first water tank is lower than the first preset liquid level, the photovoltaic cleaning drone is controlled to return to the take-off and landing point to add dissolving liquid to the first water tank.

[0113] The first preset liquid level is the alarm liquid level of the first water tank 110. When the first liquid level sensor 111 detects that the actual liquid level of the first water tank 110 is lower than the first preset liquid level, the surface dissolving liquid is insufficient and it is necessary to return to the landing point to replenish the dissolving liquid.

[0114] 2) When the second liquid level sensor detects that the actual liquid level in the second water tank is lower than the second preset liquid level, control the photovoltaic cleaning drone to return to the take-off and landing point to add cleaning fluid to the second water tank.

[0115] The second preset liquid level is the alarm liquid level of the second water tank 120. When the second liquid level sensor 121 detects that the actual liquid level of the second water tank 120 is lower than the second preset liquid level, the surface cleaning liquid is insufficient and it is necessary to return to the landing point to replenish the cleaning liquid.

[0116] This example is configured to provide a liquid level alarm function so that the dissolving and cleaning solutions can be replenished in a timely manner.

[0117] In summary, the embodiments of the present invention provide detailed descriptions of the liquid supply system components and their connections in the cleaning system, including the first water tank 110, the second water tank 120, the nozzle 131, the pump system (including the diaphragm pump 132 and the booster pump 134), the three-way solenoid valve 136, the pressure sensor 135, and the nozzle angle adjustment structure 133. Furthermore, both the independently configured first water tank 110 and the second water tank 120 are equipped with liquid level sensors, which can provide real-time feedback on the remaining liquid level in the tanks.

[0118] Specifically, regarding the spraying structure: the nozzle 131, together with the diaphragm pump 132 and the nozzle angle adjustment mechanism 133, constitutes a targeted spraying module, which is connected to the first water tank 110. The flow rate of the diaphragm pump 132 is adjustable, and the nozzle 131 is a rotatable atomizing nozzle. The nozzle angle adjustment mechanism 133 can be formed by a servo motor. According to the coordinates of the stain area output by the positioning system, the flight control system controls the drone to hover directly above the stain area. The servo motor adjusts the nozzle angle so that the dissolving liquid accurately covers the stain area. The coverage area of ​​the dissolving liquid matches the area of ​​the stain area, realizing "fixed-point and quantitative spraying". In addition, the nozzle 131, together with the pressurization pump 134 and the pressure sensor 135, constitutes a high-pressure rinsing module, which is connected to the second water tank 120. The pressurization pump 134 has adjustable pressure. During the rinsing stage, the rinsing water pressure can be intelligently adjusted according to the cleaning effect feedback from the visual recognition system or the preset program to achieve efficient rinsing and avoid damage to the photovoltaic panel.

[0119] Based on the above embodiments, optionally, the surface of the photovoltaic array includes multiple stained areas. The method includes:

[0120] 1) Obtain the stain cleaning list.

[0121] The stain cleaning list includes the cleaning order for each stained area on the photovoltaic array. This stain cleaning list can be obtained by performing a pre-scan of the photovoltaic array surface.

[0122] 2) According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spraying structure to spray the dissolving liquid onto each stain area.

[0123] In other words, according to the cleaning order provided in the stain cleaning list, the photovoltaic cleaning drone is controlled to spray the dissolving solution onto each stained area in sequence. Once all stained areas have been sprayed, the photovoltaic cleaning drone can be controlled to return to the starting stained area (i.e., the first stained area) in the list and wait for it to meet the dissolving conditions.

[0124] 3) According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spray structure to pressurize and rinse the stain area with cleaning fluid when the stain area meets the dissolution conditions.

[0125] In other words, according to the cleaning sequence provided in the stain cleaning list, the photovoltaic cleaning drone is controlled sequentially to pressurize and rinse each stained area with cleaning fluid. It is understandable that for each stained area, the rinsing is only performed after the stain has met the required dissolution conditions.

[0126] This embodiment is configured to clean each stained area according to the stain cleaning list, which can achieve precise and efficient stain removal.

[0127] Based on the above embodiments, optionally, after the dissolving solution has been sprayed on all stained areas, the method further includes:

[0128] Determine whether the battery life of the photovoltaic cleaning drone meets the preset battery life conditions.

[0129] If so, control the photovoltaic cleaning drone to fly over the starting stain area in the stain cleaning list.

[0130] If not, control the photovoltaic cleaning drone to return to the take-off and landing point for charging, and after charging is complete, fly over the starting stain area in the stain cleaning list.

[0131] The preset endurance condition is, for example, the endurance required for the photovoltaic cleaning drone to complete the rinsing of all stained areas. This embodiment, by testing the drone's endurance after the dissolving solution is sprayed, ensures that the photovoltaic cleaning drone can continuously complete the rinsing of each stained area, avoiding interruptions in cleaning operations due to limited endurance.

[0132] Understandably, when the photovoltaic cleaning drone flies over the starting stain area in the stain cleaning list, if the starting stain area has already met the dissolution conditions, there is no need to wait, and the rinsing of the starting stain area can be carried out directly.

[0133] Based on the above embodiments, optionally, the dissolving conditions include: the dissolving time after spraying the dissolving solution reaches a second preset time, and / or, the dissolving effect reaches a preset effect. Achieving the desired dissolving effect ensures that the stains are fully soaked and dissolved, and rinsing after the dissolving time reaches the second preset time avoids excessively long waiting times. Meeting either of the above two conditions is considered sufficient to meet the dissolving conditions. For example, the second preset time can be from several minutes to tens of minutes.

[0134] Specifically, the photovoltaic cleaning drone also includes a visual recognition module. Correspondingly, the method further includes: for any stained area: after spraying the dissolving solution onto the stained area and controlling the photovoltaic cleaning drone to fly over the stained area again, every third preset time interval, controlling the visual recognition module to acquire an image of the stained area, and determining the dissolving effect of the stained area through a preset dissolving effect judgment model.

[0135] The third preset time is shorter than the second preset time. For example, the third preset time is 10 seconds. The preset dissolution effect judgment model is a pre-trained model, for example, one optimized and trained based on the YOLO algorithm.

[0136] In this way, by determining the dissolution effect at each third preset time, it is possible to judge in time whether the dissolution effect meets the standard, so that rinsing can be carried out in time when the dissolution effect reaches the preset effect / the dissolution time reaches the second preset time.

[0137] Optionally, based on the above embodiments, a stain cleaning list is obtained, including:

[0138] 1) Obtain the location of each stain area on the photovoltaic array.

[0139] 2) Select any stained area as the starting stained area, and sort the stained areas according to the shortest path principle to form a stain cleaning list.

[0140] For example, a stained area at the edge of the photovoltaic array can be selected as the starting stained area, and the next stained area of ​​each stained area can be selected according to the shortest path principle, that is, according to the nearest principle, so as to form a stained cleaning list with the shortest flight path, effectively reducing the invalid flight of the drone.

[0141] In one embodiment, optionally, an inspection drone or other inspection equipment can be used to perform a comprehensive scanning inspection of the photovoltaic array surface to determine the location of each stain area and provide it to the photovoltaic cleaning drone, so that the photovoltaic cleaning drone only performs cleaning operations.

[0142] Alternatively, in another implementation, the photovoltaic cleaning drone can be equipped with a visual recognition system and a positioning system. The photovoltaic drone is first controlled to scan and inspect to determine the location of each stain area, and then a stain cleaning list is generated and the cleaning operation is carried out accordingly. In this way, the photovoltaic cleaning drone can complete the entire inspection and cleaning process without the need for other inspection equipment.

[0143] Specifically, obtaining the location of each stain area on the photovoltaic array includes: controlling the photovoltaic cleaning drone to fly along a preset route, and controlling the visual recognition system in conjunction with the positioning system to identify each stain area and its location on the surface of the photovoltaic array.

[0144] The visual recognition system can identify each stain area using a preset stain recognition model (e.g., a pre-trained model, specifically optimized and trained based on the YOLO algorithm). The positioning system can output the coordinates of each stain area, such as three-dimensional coordinates, specifically including the X / Y / Z axis positions relative to a preset zero point in the photovoltaic array. For example, when forming a stain cleaning list, the first stain area on the preset route can be selected as the starting stain area, and the cleaning order of subsequent stain areas can be determined sequentially according to the principle of proximity.

[0145] Figure 5 This is a schematic diagram of the cleaning operation path planning provided in an embodiment of the present invention. See also... Figure 5 For example, dashed lines represent preset flight paths, and solid arrows represent cleaning routes formed according to a stain cleaning list. The photovoltaic array 90 may include multiple photovoltaic panels 901 arranged in an array. The preset flight path is, for example, an S-shaped flight path traversing each photovoltaic panel 901, with the take-off and landing point 801 located, for example, at the starting position of the preset flight path. During the scanning and inspection process, the photovoltaic cleaning drone can take off from the take-off and landing point 801, traverse the area above each photovoltaic panel according to the preset flight path, and control the visual recognition system to capture images of the surface of each photovoltaic panel 901 during flight, identifying the stain areas, their area, and the degree of contamination on the surface of each photovoltaic panel 901. Simultaneously, the positioning system determines the position of each stain area and outputs its coordinates, such as the coordinates of its centroid. After completing the scanning and inspection, a stain area distribution list can be formed, marking the position of each stain area. Then, the first stain area along the preset flight path in the stain area distribution list can be used as the starting stain area. Based on the shortest path principle, the stain areas are reordered to form a stain cleaning list, for example, with... Figure 5Follow the cleaning sequence indicated by the arrows for the stained areas. By planning the cleaning path according to the "proximity principle," unnecessary flights can be reduced, improving the efficiency of the cleaning operation.

[0146] Figure 6 This is another flowchart of the cleaning control method for a photovoltaic cleaning drone provided in an embodiment of the present invention. Figure 6 The document primarily provides the photovoltaic array scanning process; see [link / reference]. Figure 6 In one specific embodiment, optionally, the cleaning control method for the photovoltaic cleaning drone includes:

[0147] S210: Control the photovoltaic cleaning drone to fly along the preset route.

[0148] S220, Control the visual recognition system to identify stained areas on the surface of the photovoltaic array.

[0149] S230. Determine if a stained area is found. If yes, proceed to S240; otherwise, return to S210.

[0150] S240, the control visual recognition system works with the positioning system to determine the area, degree of contamination and location of the stained area.

[0151] S250. Determine whether the scanning of the photovoltaic array is complete. If yes, proceed to S260; otherwise, return to S210.

[0152] S260, Generate a stain cleaning list.

[0153] S270. According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spraying structure to spray the dissolving liquid onto each stain area.

[0154] S280. According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spraying structure to pressurize and rinse the stain area with cleaning fluid when the stain area meets the dissolution conditions.

[0155] This embodiment provides a complete operation process for a photovoltaic cleaning drone through steps S210-S280. It uses visual recognition to accurately locate stains and achieves precise and quantitative spraying of the dissolving solution, thus avoiding resource waste.

[0156] Based on the above embodiments, optionally, after pressurized rinsing with cleaning fluid has been completed on all stained areas, the method further includes:

[0157] 1) If there are residual stains on the surface of the photovoltaic array, perform the steps of spraying dissolving liquid and pressurizing the cleaning liquid to rinse each residual stain area separately when the dissolution conditions are met.

[0158] The residual stain area is defined as the stain area where the difference between the image grayscale value after pressurized rinsing with cleaning fluid and the image grayscale value of the clean area is greater than a preset threshold. For example, the preset threshold is 5%.

[0159] For example, after rinsing all stained areas, a photovoltaic cleaning drone can be controlled to fly along a preset route and a visual recognition system can be controlled to perform secondary image acquisition to determine whether there are any residual stained areas on the photovoltaic array surface. The grayscale comparison between the acquired secondary image and the image of the clean area can clearly determine whether the cleaning effect of the stained areas meets the standards.

[0160] 2) If there are no residual stains on the surface of the photovoltaic array, control the photovoltaic cleaning drone to return to the take-off and landing point to complete the cleaning operation.

[0161] After the cleaning operation is completed, a cleaning report can be generated and uploaded to the ground station. The cleaning report includes information such as the stained area and the cleaning process.

[0162] Figure 7 This is another flowchart of the cleaning control method for a photovoltaic cleaning drone provided in this embodiment of the invention. Figure 7 This mainly illustrates the operation of the visual recognition system; see [link / reference]. Figure 7 In one specific embodiment, optionally, the cleaning control method for the photovoltaic cleaning drone includes:

[0163] S301. Obtain an image of the photovoltaic panel.

[0164] For example, a visual recognition system may specifically include a camera and an image processing unit, with the image processing unit connected to both the camera and the control unit. The camera may be a high-definition industrial camera, and the image processing unit may be an AI image processing unit, equipped with stain recognition algorithms and cleanliness assessment algorithms. For instance, it may be configured with preset stain recognition models, preset dissolution effect judgment models, and preset cleanliness assessment models. These models can all be optimized based on the YOLO algorithm, for example, trained on the morphological, texture, and color features of bird droppings and dried stains. The degree of contamination can be specifically classified into three levels: light contamination, moderate contamination, and heavy contamination, based on factors such as stain coverage area and viscosity level.

[0165] S302, Image preprocessing.

[0166] Specifically, preprocessing may include image normalization, image noise reduction, and image enhancement.

[0167] S303. Identify stain areas in photovoltaic panel images based on a preset stain recognition model.

[0168] S304. Extract features from the stained area.

[0169] Specifically, features such as the shape, color, and texture of the stained area can be extracted, and the degree of contamination in the stained area can be determined accordingly.

[0170] S305, Lightly polluted.

[0171] S306, moderate pollution.

[0172] S307, Severely polluted.

[0173] S308, Generate spraying and rinsing parameters for the stained area.

[0174] S309. Spray the dissolving solution onto the stained area according to the spraying parameters.

[0175] S310. When the stain area meets the dissolution conditions, the stain area is pressurized and rinsed with cleaning fluid according to the rinsing parameters.

[0176] S311. Determine if the cleanliness of the stained area meets the standard? If yes, proceed to S312; otherwise, proceed to S313.

[0177] S312, Cleaning complete.

[0178] S313. Reassess the degree of contamination in the stained area.

[0179] Specifically, the operation of S313 can repeat the steps of S301-S307. S308-S313 can all be considered as a cleaning process for the same stained area, and this cleaning process can be used for any stained area.

[0180] This embodiment provides a complete operation process for a photovoltaic cleaning drone based on S301-S313, mainly improving stain identification and cleanliness assessment and related processes, and demonstrating the algorithm process based on YOLO for stain area identification, feature extraction, pollution degree classification, cleaning parameter generation and cleanliness assessment.

[0181] Figure 8 This is another flowchart of the cleaning control method for the photovoltaic cleaning drone provided in this embodiment of the invention. See also... Figure 8 In one embodiment, optionally, the cleaning control method for the photovoltaic cleaning drone includes:

[0182] S410: Control the photovoltaic cleaning drone to fly along a preset route, control the visual recognition system in conjunction with the positioning system to identify the area, location and degree of contamination of each stain area on the surface of the photovoltaic array, and generate a stain cleaning list.

[0183] S420: According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spraying structure to spray the dissolving liquid onto each stain area.

[0184] S430: Control the photovoltaic cleaning drone to fly over the starting stain area in the stain cleaning list.

[0185] S440. According to the stain cleaning list, control the photovoltaic cleaning drone to fly over each stain area in sequence, and control the spray structure to pressurize and rinse the stain area with cleaning fluid when the stain area meets the dissolution conditions.

[0186] S450. In the case of residual stains on the surface of the photovoltaic array, perform the steps of spraying dissolving liquid and pressurizing the cleaning liquid to rinse each residual stain area separately when the dissolution conditions are met.

[0187] S460: When there are no residual stains on the surface of the photovoltaic array, control the photovoltaic cleaning drone to return to the take-off and landing point to complete the cleaning operation.

[0188] This embodiment provides a complete operation process for a photovoltaic cleaning drone based on S410-S460. Addressing the problems of incomplete removal of stubborn stains, high resource consumption, and lack of targeted treatment capabilities in current photovoltaic cleaning drones, it provides a dual-water-tank fixed-point targeted photovoltaic cleaning drone and cleaning method. Through "dual-water-tank functional zoning + visual targeted positioning + step-by-step processing", it achieves efficient and precise removal of stubborn stains, while reducing the consumption of dissolving and cleaning solutions.

[0189] Figure 9 These are comparison images of the cleaning effects of the photovoltaic cleaning drone provided in this embodiment of the invention. (See attached image) Figure 9 The diagram uses squares to represent photovoltaic arrays and dashed lines to represent stubborn stains. The stain coverage rate represents the percentage of stubborn stains covering the entire array. It's evident that the stain coverage was high before cleaning. Analysis shows that when the stain coverage reaches 35%, the power generation efficiency drops to the nominal 72%. After cleaning, the stain coverage can be reduced to less than 2%, increasing the power generation efficiency to 98%. Compared to traditional cleaning methods, which achieve a 70% stain removal rate but use 5L / m² of water... 2 The method described in this application achieves a stain removal rate of 95% while using only 1.2 L / m³ of water. 2 The removal rate was increased by 35.7%, while water consumption decreased by 76%, achieving precise stain removal and water conservation. It is understood that the method provided in this embodiment of the invention primarily targets stubborn stains. This photovoltaic cleaning drone and its cleaning control method can complement ordinary cleaning drones. For example, ordinary cleaning drones utilize large-area airflow or water flow to clean ordinary dust and simple attachments (such as easily removable stains like paper or fallen leaves), while stubborn stains that ordinary cleaning drones cannot remove are cleaned by this photovoltaic cleaning drone.

[0190] In summary, the cleaning control method provided by the embodiments of the present invention can achieve precise, efficient, and water-saving cleaning of photovoltaic panels, especially for the effective removal of stubborn stains such as bird droppings and mud. It provides a step-by-step treatment mechanism of "dissolving-targeted rinsing" to achieve the synergistic operation of "small-dose precise dissolution + medium-high pressure targeted rinsing", which effectively improves the cleaning effect and resource utilization rate for stubborn stains.

[0191] Specifically, this photovoltaic cleaning drone features a dual-tank independent liquid supply system, effectively preventing the mixing of the dissolving solution and cleaning solution. Switching between the supply tanks according to different cleaning stages improves cleaning solution utilization. A visual recognition system enables precise, point-to-point spraying of both cleaning and dissolving solutions, ensuring high accuracy in locating the center of the stain area and avoiding waste from large-area spraying. It is highly targeted at specific stains, resulting in excellent cleaning performance. It also features intelligent stain recognition; deep learning-based stain type identification effectively classifies the degree of contamination (light, moderate, heavy) and automatically generates the optimal cleaning strategy accordingly. Furthermore, it has an adaptive pressure adjustment function, automatically adjusting the spray pressure based on stain type, with low pressure for dissolving solution spraying and high pressure for cleaning solution rinsing. The positioning system utilizes GPS / RTK positioning, and by incorporating a high-precision positioning module (such as GPS + BeiDou dual-mode positioning) and obstacle avoidance sensors, the flight control system supports centimeter-level hovering, ensuring the drone accurately locates stains above the photovoltaic panels. Therefore, this photovoltaic cleaning drone possesses the following advantages:

[0192] 1. Water and energy saving: Compared with traditional large-area spraying, water consumption is reduced by more than 70%.

[0193] 2. Excellent cleaning effect: The two-step method of dissolving and rinsing can remove stubborn stains with a removal rate of over 95%.

[0194] 3. High efficiency: Automated operation, planning the stain treatment path according to the "nearest principle", combined with centimeter-level fixed-point hovering, to avoid ineffective flight.

[0195] 4. High level of intelligence: The entire process of automatic identification, positioning, cleaning, and verification is intelligent.

[0196] 5. Low cost: Reduces labor costs and lowers water consumption.

[0197] This invention also provides a photovoltaic cleaning drone, which can be used to clean the surface of a photovoltaic array by applying the cleaning control method of the photovoltaic cleaning drone provided in any embodiment of this invention, and has corresponding beneficial effects.

[0198] See Figure 1The photovoltaic cleaning drone includes a cleaning system and a control unit 200. The cleaning system includes a first water tank 110, a second water tank 120, and a spraying structure 130. The first water tank 110 stores a dissolving solution, and the second water tank 120 stores a cleaning solution. The first water tank 110 and the second water tank 120 are respectively connected to the spraying structure 130. The control unit 200 is used to execute the cleaning control method of the photovoltaic cleaning drone provided in any embodiment of the present invention.

[0199] For example, the photovoltaic cleaning drone may also include a flight control system, a vision recognition system 300, and a positioning system 400. The flight control system may include, for example, a flight platform 500, which may specifically be a multi-rotor flight platform. Figure 1 The main exhibit showcases the overall structure of the drone, including the layout of its core components. The control unit 200, as the core processing component of the photovoltaic cleaning drone, can control the flight control system, cleaning system, visual recognition system 300, and positioning system 400 respectively.

[0200] Based on the above embodiments, optionally, the photovoltaic cleaning drone also includes a wireless communication module connected to the control unit 200. The control unit 200 can communicate with a ground station via the wireless communication module. The ground station can act as a host computer, enabling task planning and real-time monitoring of the photovoltaic cleaning drone. The wireless communication module can achieve 2.4G / 5.8G wireless communication.

[0201] This invention also provides a photovoltaic cleaning system, including a ground station and a photovoltaic cleaning drone provided in any embodiment of this invention, which has corresponding beneficial effects.

[0202] Figure 10 This is a schematic diagram of the photovoltaic cleaning system provided in an embodiment of the present invention. See also... Figure 10 The photovoltaic cleaning system may include a ground station 2000 and at least one photovoltaic cleaning drone 1000. The photovoltaic cleaning drone 1000 may include a cleaning system 100, a control unit 200, a vision recognition system 300, a positioning system 400, and a flight system 600. The control unit 200 is connected to the cleaning system 100, the vision recognition system 300, the positioning system 400, and the flight system 600, and communicates with the ground station 2000 via a wireless communication module. The vision recognition system 300 can transmit stain data to the control unit 200, and the positioning system 400 can transmit location data to the control unit 200. It is understood that sensors in each functional system, such as pressure / liquid level / obstacle avoidance sensors, are also connected to the control unit 200 to provide real-time feedback of relevant status information. The control unit 200 can issue control commands to the corresponding functional systems based on the feedback information.

[0203] All control functions of the photovoltaic cleaning drone 1000 are integrated into the control unit 200, which interacts with the ground station 2000. The ground station 2000 is used for task planning, monitoring operation status, and manual intervention, supporting remote monitoring and intervention. For example, operators can view real-time images and cleaning progress transmitted by the drone on the ground station 2000 and make adjustments as necessary.

[0204] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0205] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A cleaning control method of a photovoltaic cleaning drone, characterized in that, The photovoltaic cleaning unmanned aerial vehicle comprises a cleaning system, the cleaning system comprising a first water tank, a second water tank and a spraying structure, the first water tank storing a dissolving solution, the second water tank storing a cleaning solution, and the first water tank and the second water tank being respectively communicated with the spraying structure; The method comprises the following steps: controlling the photovoltaic cleaning unmanned aerial vehicle to fly above the stain area, and controlling the spraying structure to spray the dissolving solution to the stain area; wherein, during the dissolving solution spraying process, the spraying parameter of the spraying structure spraying the dissolving solution is determined according to the area and the pollution degree of the stain area; when the stain area meets the dissolving condition, controlling the spraying structure to perform the cleaning solution pressurized flushing to the stain area; wherein, during the flushing process, the flushing parameter of the spraying structure performing the cleaning solution pressurized flushing is adjusted according to the area and the pollution degree of the stain area. 2.The cleaning control method of the photovoltaic cleaning drone according to claim 1, wherein, The spraying structure comprises a diaphragm pump, a nozzle and a nozzle angle adjusting mechanism; the first water tank is communicated with the nozzle through the diaphragm pump; determining the spraying parameter of the spraying structure spraying the dissolving solution according to the area and the pollution degree of the stain area comprises: adjusting the flow of the diaphragm pump according to the area of the stain area; controlling the nozzle angle adjusting mechanism to adjust the angle of the nozzle according to the relative position of the nozzle and the stain area, so that the dissolving solution sprayed by the nozzle covers the stain area; adjusting the spraying time of the nozzle spraying the dissolving solution according to the pollution degree of the stain area; wherein, the higher the pollution degree is, the longer the spraying time is. 3.The cleaning control method of the photovoltaic cleaning drone according to claim 1, wherein, The spraying structure comprises a pressurized pump and a nozzle; the second water tank is communicated with the nozzle through the pressurized pump; adjusting the flushing parameter of the spraying structure performing the cleaning solution pressurized flushing according to the area and the pollution degree of the stain area comprises: adjusting the caliber of the nozzle according to the area of the stain area; adjusting the flushing water pressure loaded to the nozzle by the pressurized pump and / or adjusting the flushing time of the nozzle according to the pollution degree of the stain area. 4.The cleaning control method of the photovoltaic cleaning drone according to claim 3, characterized in that, The photovoltaic cleaning unmanned aerial vehicle further comprises a visual identification system; adjusting the flushing water pressure loaded to the nozzle by the pressurized pump and / or adjusting the flushing time of the nozzle according to the pollution degree of the stain area comprises: determining the lower limit flushing water pressure, the upper limit flushing water pressure and the maximum flushing time according to the pollution degree of the stain area; controlling the pressurized pump to provide the lower limit water pressure and to open the nozzle to spray the cleaning solution to the stain area, so as to start the flushing to the stain area; During the flushing process, every first preset time interval, the visual recognition system is controlled to collect an image of the stain area and compare it with the image of the stain area before flushing, and obtain an actual stain removal effect based on a preset cleanliness evaluation model; in the case where the actual stain removal effect does not reach a preset stain removal effect, the current flushing water pressure is used to continue flushing, or the flushing water pressure is increased to continue cleaning, until the actual stain removal effect reaches the preset stain removal effect or the flushing time reaches the maximum flushing time, and the flushing of the stain area is stopped; wherein the flushing water pressure used during the flushing process does not exceed the upper limit of the flushing water pressure. 5.The cleaning control method of the photovoltaic cleaning drone according to claim 4, characterized in that, The spraying structure further comprises a pressure sensor for detecting the water pressure of the nozzle; the method further comprises: During the flushing process, the actual water pressure of the nozzle detected by the pressure sensor is fed back to adjust the flushing water pressure provided by the pressure pump.

6. The cleaning control method of the photovoltaic cleaning drone according to any one of claims 2-5, characterized in that, The spraying structure further comprises a three-way electromagnetic valve, and the first water tank and the second water tank are connected to the nozzle through the three-way electromagnetic valve; the method further comprises: When the dissolved solution needs to be sprayed, the three-way electromagnetic valve is controlled to connect the first water tank and the nozzle; When the cleaning liquid needs to be pressurized and flushed, the three-way electromagnetic valve is controlled to connect the second water tank and the nozzle.

7. The cleaning control method of the photovoltaic cleaning drone according to any one of claims 2-5, characterized in that, The cleaning system further comprises a first liquid level sensor and a second liquid level sensor for detecting the liquid level of the first water tank and the liquid level of the second water tank respectively; the method further comprises: When the first liquid level sensor detects that the actual liquid level of the first water tank is lower than a first preset liquid level, the photovoltaic cleaning unmanned aerial vehicle is controlled to return to the take-off point to add dissolved solution to the first water tank; When the second liquid level sensor detects that the actual liquid level of the second water tank is lower than a second preset liquid level, the photovoltaic cleaning unmanned aerial vehicle is controlled to return to the take-off point to add cleaning liquid to the second water tank. 8.The cleaning control method of the photovoltaic cleaning drone according to claim 1, wherein, The surface of the photovoltaic array comprises a plurality of stain areas; the method comprises: obtaining a stain cleaning list; wherein the stain cleaning list comprises the cleaning sequence of each stain area on the photovoltaic array; According to the stain cleaning list, the photovoltaic cleaning unmanned aerial vehicle is controlled to fly over each stain area in turn, and the spraying structure is controlled to spray dissolved solution to each stain area; According to the stain cleaning list, the photovoltaic cleaning unmanned aerial vehicle is controlled to fly over each stain area in turn, and the spraying structure is controlled to pressurize and flush the stain area with cleaning liquid if the stain area meets the dissolution condition. 9.The cleaning control method of the photovoltaic cleaning drone according to claim 1 or 8, characterized in that, The dissolution condition includes that the dissolution time after spraying the dissolved solution reaches a second preset time, and / or the dissolution effect reaches a preset effect. 10.The cleaning control method of the photovoltaic cleaning drone according to claim 9, wherein, The photovoltaic cleaning unmanned aerial vehicle further comprises a visual recognition module; The method further comprises, for any stain area: After the spraying of the dissolving solution to the stain area is completed and the photovoltaic cleaning unmanned aerial vehicle is controlled to fly over the stain area again, every third preset time, the visual recognition module is controlled to collect an image of the stain area once, and the dissolving effect of the stain area is determined through the preset dissolving effect determination model; wherein the third preset time is less than the second preset time. 11.The cleaning control method of the photovoltaic cleaning drone according to claim 8, wherein, The stain cleaning list is obtained by: Obtaining the positions of each stain area on the photovoltaic array; Selecting any stain area as a starting stain area, and sorting each stain area according to the shortest path principle to form the stain cleaning list. 12.The cleaning control method of the photovoltaic cleaning drone according to claim 11, wherein, The photovoltaic cleaning unmanned aerial vehicle includes a visual recognition system and a positioning system. Obtaining the positions of each stain area on the photovoltaic array includes: Controlling the photovoltaic cleaning unmanned aerial vehicle to fly according to a preset flight path, and controlling the visual recognition system to identify each stain area and its position on the surface of the photovoltaic array in cooperation with the positioning system; Wherein, the first stain area on the preset flight path is selected as the starting stain area. 13.The cleaning control method of the photovoltaic cleaning drone according to claim 8, wherein, After the spraying of the dissolving solution to all the stain areas is completed, the method further includes: Determining whether the endurance capability of the photovoltaic cleaning unmanned aerial vehicle meets a preset endurance condition; If yes, controlling the photovoltaic cleaning unmanned aerial vehicle to fly to the starting stain area in the stain cleaning list; If no, controlling the photovoltaic cleaning unmanned aerial vehicle to return to the take-off point for charging, and flying to the starting stain area in the stain cleaning list after the charging is completed. 14.The cleaning control method of the photovoltaic cleaning drone according to claim 1 or 8, characterized in that, After the pressurized washing of the cleaning liquid to all the stain areas is completed, the method further includes: In the case that there are residual stain areas on the surface of the photovoltaic array, the steps of spraying a dissolving solution and pressurized washing of a cleaning liquid when the dissolving condition is met are performed for each residual stain area respectively; wherein the residual stain area is a stain area whose image gray scale difference after the pressurized washing of the cleaning liquid and the image gray scale of a clean area is greater than a preset threshold value; In the case that there is no residual stain area on the surface of the photovoltaic array, controlling the photovoltaic cleaning unmanned aerial vehicle to return to the take-off point to complete the cleaning operation.

15. A photovoltaic cleaning drone, characterized in that, It includes: a cleaning system and a control unit; The cleaning system includes a first water tank, a second water tank and a spraying structure, the first water tank stores a dissolving solution, the second water tank stores a cleaning liquid, and the first water tank and the second water tank are respectively connected to the spraying structure; the control unit is used to execute the cleaning control method of the photovoltaic cleaning unmanned aerial vehicle according to any one of claims 1-14.

16. The photovoltaic cleaning drone of claim 15, wherein, It further includes: a wireless communication module connected to the control unit; wherein the control unit communicates with the ground station through the wireless communication module.

17. A photovoltaic cleaning system, characterized by It includes: a ground station and a photovoltaic cleaning unmanned aerial vehicle as claimed in claim 15 or 16.

Images