Mountain / hill photovoltaic power station large-area cleaning robot

Abstract

The invention provides a cleaning robot for a large area of a mountain / hill photovoltaic power station, and aims to solve the problems that traditional equipment is low in cleaning efficiency and limited in operation range under uneven terrains and assembly dislocation scenes. The robot comprises a track laid on a photovoltaic array, a portal frame moving along the track through a bottom driving wheel, a hydraulic cylinder used for adjusting the levelness of a cross beam, and a tilt angle sensor (monitoring the state of the cross beam and triggering the hydraulic cylinder to adjust). The mechanical arm is installed on a cross beam of the portal frame and can move, the tail end of the mechanical arm is provided with a cleaning detection assembly (comprising a camera, a nozzle and a brush), the camera collects images of the photovoltaic panel and establishes a dynamic pollution level database based on a pollution level function F (xi, yi, zi, t), and the nozzle and the brush are combined with water washing and brushing to clean the photovoltaic panel. The controller cooperatively controls the moving assembly, the mechanical arm and the cleaning detection assembly, optimizes differentiated cleaning strategies by analyzing a dynamic database, and judges whether water-assisted cleaning is started or not and the cleaning frequency according to a pollution level function.

Description

Technical Field

[0001] The present invention relates to a large-area cleaning robot for mountain / hill photovoltaic power stations. Background Art

[0002] As an important tool for improving power generation efficiency, the photovoltaic panel cleaning robot still faces multiple practical challenges in its large-scale application despite continuous technological iteration. First of all, the problem of insufficient adaptability to complex environments is particularly prominent. The robot has strict requirements for the installation accuracy of the photovoltaic array and the terrain conditions. When the misalignment of the photovoltaic panel exceeds 50 mm or the slope exceeds 25 degrees, the moving stability and cleaning coverage will significantly decrease. This technical limitation has led some mountain photovoltaic power stations to abandon robot operations due to the severe terrain undulation and instead adopt manual cleaning methods.

[0003] Secondly, the complexity of equipment maintenance severely restricts its popularity. The models using negative pressure adsorption technology generally have a failure mode of air leakage caused by the wear of the suction cups, and vulnerable parts need to be replaced regularly to maintain the sealing performance; for faults involving sensor networks and intelligent control systems, professional maintenance teams often need to intervene for repair. For large photovoltaic power stations located in remote areas, this maintenance mode relying on professional technical personnel may lead to a lag in response and affect the equipment availability.

[0004] At the technical level, the contradiction between cleaning efficiency and energy consumption is becoming increasingly prominent. Although the track-type robot has the advantage of fast operation, it is limited by the fixed track layout and is difficult to meet the needs of distributed photovoltaic scenarios; while the autonomous mobile robot has good spatial adaptability, but faces inherent defects such as slow moving speed and high energy consumption. Especially the models equipped with vacuum adsorption devices have insufficient endurance in continuous working conditions and need to be charged frequently or rely on external power sources, which forms a new application bottleneck in remote areas lacking stable power supply. Summary of the Invention

[0005] The present invention provides a large-area cleaning robot for mountain / hill photovoltaic power stations, which can effectively solve the technical bottlenecks such as low cleaning efficiency and limited operation range of traditional cleaning equipment in non-flat terrain and photovoltaic module misalignment scenarios.

[0006] A large - area cleaning robot for mountain / hill photovoltaic power stations, comprising: a moving component, the moving component includes: a track, laid on the installation site of the photovoltaic array; a gantry, installed on the track and moving along the track through a bottom driving wheel, a pulley is arranged on the crossbeam of the gantry and a stepping motor for driving the pulley to move along the crossbeam; a driving motor, configured to drive the driving wheel to drive the whole gantry to move; a hydraulic cylinder, located at the top of the gantry column, used to adjust the levelness of the crossbeam; and an inclination sensor, configured to monitor the horizontal state of the crossbeam and trigger the hydraulic cylinder to adjust when detecting that the crossbeam is not horizontal; a robotic arm, installed on the pulley; a cleaning detection component, installed at the end of the robotic arm, including a camera, a nozzle and a brush, wherein the camera is used to collect the surface image of the photovoltaic panel and establish a dynamic pollution level database based on a preset pollution level function F(x i ,y i ,z i ,t), the nozzle and the brush clean the photovoltaic panel by combining two methods of water flow scouring and brush cleaning; and a controller, configured to control the coordinated actions of the moving component, the cleaning detection component and the robotic arm, analyze the pollution accumulation law of each photovoltaic panel based on the dynamic pollution level database, optimize the differential cleaning strategy, and determine whether to start water - assisted cleaning and the cleaning frequency according to the pollution level function F(x i ,y i ,z i ,t).

[0007] In some examples, the dynamic pollution level database at least includes the following fields: photovoltaic panel ID; coordinate position (x i ,y i ,z i ); cleaning time t; the value of the pollution level function F(x i ,y i ,z i ,t).

[0008] In some examples, the controller controls the moving component, the cleaning detection component and the robotic arm to achieve automatic cleaning through the following steps: the driving motor controls the gantry to move along the track to the target photovoltaic panel coordinate position; the inclination sensor detects the levelness of the crossbeam and triggers the hydraulic cylinder to adjust; the stepping motor drives the pulley to move to the center of the target photovoltaic panel, the robotic arm adjusts the orientation of the cleaning detection component, and selects to start the brush and / or the nozzle to clean the target photovoltaic panel based on the pollution level function F(x i ,y i ,z i ,t) in the dynamic pollution level database; after the cleaning is completed, the robotic arm resets and returns to the initial position.

[0009] In some examples, the pollution level function F(x i ,y i ,z i ,t) ∈ [0, 5], where: Level 0: The surface of the photovoltaic panel is pollution-free, and the reflectivity of the photovoltaic panel ≥ 95%; Level 1: The decrease in the reflectivity of the photovoltaic panel ≤ 5%, and the area covered by deposited dust ≤ 5%; Level 2: The decrease in the reflectivity of the photovoltaic panel is 5% - 15%, and the area covered by deposited dust ≥ 5% and ≤ 15%; Level 3: The decrease in the reflectivity of the photovoltaic panel is 15% - 30%, and moss and lichen appear at the edges; Level 4: The decrease in the reflectivity of the photovoltaic panel is 30% - 50%, the area covered by bird droppings ≥ 10%, or moss and lichen spread along the edges; Level 5: The decrease in the reflectivity of the photovoltaic panel ≥ 50%, the area covered by bird droppings ≥ 30%, or moss and lichen spread from the edges to the center.

[0010] In some examples, the dynamic pollution level database collects pre-cleaning images through the camera, and evaluates the pollution level of the target photovoltaic panel based on the pollution level function F(x i ,y i ,z i ,t); if the pollution level of the target photovoltaic panel ≥ 2, the nozzle is activated for water-assisted cleaning.

[0011] In some examples, the method for adjusting the horizontal level of the crossbeam includes: the tilt sensor monitors the horizontal state of the crossbeam in real time and transmits the data to the controller; the controller controls the telescopic movement of the hydraulic cylinder according to the deviation value feedback by the tilt sensor to adjust the horizontal level of the crossbeam.

[0012] In some examples, the method for determining the cleaning frequency includes: defining a pollution increment function G1(dt) = F(x i ,y i ,z i ,t1 + dt) - F(x i ,y i ,z i ,t1), observing the change of G1(dt) by adjusting the cleaning interval time dt: initially set dt_min and dt_max as the preset minimum and maximum cleaning interval times; within the range of dt_min to dt_max, gradually adjust dt and calculate G1(dt); if G1(dt) ∈ (0, 2], the current dt is the optimal cleaning interval time; if G1(dt) > 2, shorten dt and re-evaluate; if G1(dt) = 0, extend dt and re-evaluate; by iterating the above steps, finally determine the dt value that makes G1(dt) stable within the range of (0, 2] as the periodic cleaning frequency. Description of the Drawings

[0013] Figure 1 It is a schematic diagram of a large - area cleaning robot for mountain / hill photovoltaic power stations in an embodiment of the present invention.

[0014] Figure 2 is Figure 1 A partial enlarged view of the robotic arm of the shown robot.

[0015] Figure 3 It is a partial enlarged view of the moving component of the partial enlarged view of the robotic arm. Detailed implementation manners

[0016] Please refer to Figures 1 to 3 , a large - area cleaning robot for mountain / hill photovoltaic power stations, comprising a moving component 100, a cleaning and detection component 200, a robotic arm 300 and a controller.

[0017] The moving component 100 includes a gantry 101, a track 104, a driving wheel 105, a driving motor 106, a hydraulic cylinder 108 and an inclination sensor 109. The track 104 is laid on the photovoltaic array installation site. The gantry 101 moves along the track 104 through the driving wheel 105 at the bottom. The driving motor 106 (such as a DC servo motor) drives the driving wheel 105 to drive the whole gantry to move. The level of the cross - beam of the gantry 101 can be dynamically adjusted by the hydraulic cylinder 108 at the top of the column. The inclination sensor 109 installed on the cross - beam monitors the horizontal state in real time. When the inclination of the cross - beam exceeds the set value (for example, 3 degrees), the hydraulic cylinders 108 at both ends automatically adjust to make the cross - beam return to horizontal, ensuring that the cleaning operation is not affected by the height difference of the terrain on both sides of the track. The gantry 101 adopts a double - column steel structure design and is equipped with a magnetic rail brake at the bottom for emergency braking. The surface of the track 104 is provided with anti - slip patterns.

[0018] A pulley 107 and a stepper motor 102 are arranged on the cross - beam. The pulley 107 bears the six - degree - of - freedom robotic arm 300. The robotic arm 300 realizes the three - dimensional space positioning of the brush 203 and the nozzle 202 through the coordinated drive of the servo motor 301, the first motor 302 and the second motor 303. The cleaning and detection component 200 is installed at the end of the robotic arm and includes a camera 201, a nozzle 202 and a brush 203.

[0019] The cleaning and detection component 200 combines two cleaning methods of water flow flushing and brush cleaning to improve the cleaning efficiency. The camera 201 can collect the surface images of the photovoltaic panels in real time. Based on the pollution level function F(x i ,y i ,z i ,t) ∈ [0, 5] (0 = no pollution, 5 = severe pollution), a pollution level database is established. By analyzing the pollution accumulation law of each photovoltaic panel, a differentiated cleaning strategy is optimized.

[0020] The pollution level database includes the following fields: PV panel ID; coordinate position (x i ,y i ,z i ); Cleaning time t; Pollution level F(x i ,y i ,z i ,t); historical cleaning records; cleaning effect comparison data (difference in reflectivity before and after cleaning).

[0021] Pollution level function F(x i ,y i ,z i ,t)∈[0,5] (0=no pollution, 5=heavy pollution) is used to quantify the degree of dirtiness on the surface of photovoltaic panels. The exemplary grading standards are as follows: Level 0: Qualitative description: There is no obvious pollution on the surface of the photovoltaic panel, and there are no particles or attachments that affect the functionality; quantitative indicators: reflectivity ≥ 95%, and the area covered by deposited dust ≤ 1%.

[0022] Level 1: Qualitative description: There is a very light layer of dust on the surface of the photovoltaic panel, which is difficult to detect with the naked eye and requires the use of a magnifying glass to observe the particles; Quantitative indicators: reflectivity decreases ≤5%, and the area covered by deposited dust ≤5%.

[0023] Level 2: Qualitative description: The deposited dust layer is visible on the photovoltaic panel, forming an obvious cover; Quantitative indicators: The reflectivity decreases by 5%-15%, and the deposited dust coverage area is ≥5% and ≤15%.

[0024] Level 3: Qualitative description: Dust deposition is stable, accompanied by traces of bird droppings, and signs of moss and lichen appear on the edges; Quantitative indicators: reflectivity decreases by 15%-30%, traces of bird droppings are visible (diameter ≥ 2mm), and the moss and lichen coverage area on the edges is ≤5%.

[0025] Level 4: There is a large amount of dirt and bird droppings on the surface of the photovoltaic panel, and moss and lichens spread along the edges; quantitative indicators: reflectivity decreases by 30%-50%, bird droppings cover area ≥10%, moss and lichens spread along the edges (cover area 10%-30%).

[0026] Level 5: Qualitative description: The surface of the photovoltaic panel is covered with a large amount of dirt and bird excrement, and moss and lichens are widely spread from the edge to the center; quantitative indicators: reflectivity decreases by ≥50%, bird excrement coverage area ≥30%, moss and lichens spread from the edge to the center (coverage area ≥50%).

[0027] Note: The above classification and corresponding descriptions are for illustrative purposes only. In practice, the classification standards and judgment basis can be adjusted according to specific needs. Calculating the surface reflectivity of photovoltaic panels based on the grayscale values of images collected by cameras is a prior art.

[0028] Taking a certain photovoltaic panel as an example, its level before the first cleaning is F(x1, y1, z1, t0). If F(x1, y1, z1, t0) ≥ 2, it is determined that water-assisted cleaning needs to be started. The level after cleaning is F(x1, y1, z1, t1). After a time interval of dt, the level before the second cleaning is F(x1, y1, z1, t1 + dt).

[0029] Define the pollution increment function G1(dt) = F(x i , y i , z i , t1 + dt) - F(x i , y i , z i , t1). Observe the change of G1(dt) by adjusting the cleaning interval time dt: Initially set dt_min and dt_max as the preset minimum and maximum cleaning interval times; Within the range of dt_min to dt_max, gradually adjust dt and calculate G1(dt); When dt gradually decreases (i.e., the cleaning frequency increases), if G1(dt) ∈ (0, 2], then the current dt is the optimal cleaning interval time. At this time, the pollution accumulation rate is moderate, which can avoid excessive cleaning and ensure the cleaning effect; If G1(dt) > 2, it means that the cleaning interval is too long and dt needs to be shortened and re-evaluated; If G1(dt) = 0, it indicates that there is no pollution accumulation or excessive cleaning, and dt needs to be extended and re-evaluated; By iterating the above steps, finally determine the dt value that makes G1(dt) stable within the range of (0, 2] as the periodic cleaning frequency of this photovoltaic panel.

[0030] The controller can be a PLC and is integrated in the control box 103 of the gantry 101. The controller controls the moving component 100, the cleaning and detection component 200 and the robotic arm 300 to achieve automatic cleaning through the following steps: First, the drive motor 106 controls the gantry 101 to move along the track 104 to the target photovoltaic panel coordinate position, and the inclination sensor 109 detects the levelness of the crossbeam and triggers the hydraulic cylinder 108 to adjust; Subsequently, the stepper motor 102 drives the pulley 107 to move to the center of the photovoltaic panel, and the robotic arm 300 adjusts the orientation of the brush 203 through the cooperation of multi-axis motors; The camera 201 collects the image before cleaning and evaluates the pollution level. If F(x, y, z, t) ≥ 2, the nozzle 202 is activated for water-assisted cleaning, and the image is collected again after cleaning to compare the effect; After cleaning a single photovoltaic panel, the robotic arm 300 resets and returns to the initial position. During the continuous operation process, the database records the pollution change trend of each photovoltaic panel, and the optimal cleaning cycle is determined by calculating G1(dt) = F(x1, y1, z1, t1 + dt) - F(x1, y1, z1, t1). When 0 < G1(dt) ≤ 2, it is determined as a stable cycle.

[0031] In summary, the robot of the present invention has high flexibility and wide coverage ability, can adapt to complex terrain environments, and can efficiently meet the high-frequency cleaning requirements of large-scale photovoltaic power stations. The device adopts the cooperative cleaning technology of water flushing and brushes, with remarkable cleaning effect, and overall realizes highly automated operation.

Claims

1. A large area cleaning robot for mountain / hilly photovoltaic power stations, characterized by: include: A mobile component, the mobile component comprising: Tracks, laid at the photovoltaic array installation site; A gantry is mounted on the track and moves along the track via a driving wheel at the bottom. A pulley and a stepper motor driving the pulley to move along the beam are provided on the beam of the gantry. A driving motor is configured to drive the driving wheel to drive the entire gantry to move; A hydraulic cylinder, located at the top of the gantry column, for adjusting the horizontality of the beam; and a tilt sensor configured to monitor the horizontal state of the beam and trigger the hydraulic cylinder to adjust when detecting that the beam is not horizontal; a robotic arm mounted on the pulley; The cleaning detection component is installed at the end of the robot arm and includes a camera, a nozzle and a brush, wherein the camera is used to collect the surface image of the photovoltaic panel and detect the pollution level based on the preset pollution level function F(x i ,y i ,z i ,t) establishing a dynamic pollution level database, the nozzle and the brush are used to clean the photovoltaic panel in a combination of water flushing and brush cleaning; and The controller is configured to control the coordinated actions of the mobile component, the cleaning detection component and the robotic arm, analyze the pollution accumulation law of each photovoltaic panel based on the dynamic pollution level database, optimize the differentiated cleaning strategy, and i ,y i ,z i ,t) Determine whether to enable water-assisted cleaning and the cleaning frequency.

2. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 1, characterized in that: The dynamic pollution level database includes at least the following fields: photovoltaic panel ID; coordinate position (x i ,y i ,z i ); Cleaning time t; Pollution level function F(x i ,y i ,z i ,t) value.

3. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 2, characterized in that: The controller controls the moving component, the cleaning detection component and the robotic arm to achieve automated cleaning through the following steps: The driving motor controls the gantry to move along the track to the target photovoltaic panel coordinate position; The tilt sensor detects the horizontality of the beam and triggers the adjustment of the hydraulic cylinder; The stepper motor drives the pulley to move to the center of the target photovoltaic panel. The robot arm adjusts the position of the cleaning detection component and detects the position of the cleaning detection component based on the pollution level function F(x i ,y i ,z i ,t) selecting and starting the brush and / or the nozzle to clean the target photovoltaic panel; After cleaning is completed, the robot arm resets and returns to its initial position.

4. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 3, characterized in that: The pollution level function F(x i ,y i ,z i ,t)∈[0,5], where: Level 0: The photovoltaic panel surface is unpolluted and the photovoltaic panel reflectivity is ≥95%; Level 1: The reflectivity of the photovoltaic panel decreases by ≤5%, and the area covered by deposited dust is ≤5%; Level 2: The reflectivity of the photovoltaic panel decreases by 5%-15%, and the area covered by deposited dust is ≥5% and ≤15%; Level 3: The reflectivity of the photovoltaic panel decreases by 15%-30%, and moss and lichen appear on the edges; Level 4: The reflectivity of the photovoltaic panel decreases by 30%-50%, bird droppings cover ≥10% of the area, or moss and lichen spread along the edge; Level 5: The reflectivity of the photovoltaic panel decreases by ≥50%, the bird droppings cover ≥30% of the area, or mosses and lichens spread from the edge to the center.

5. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 4, characterized in that: The dynamic pollution level database collects the image before cleaning through the camera and uses the pollution level function F(x i ,y i ,z i ,t) Evaluate the pollution level of the target photovoltaic panel; if the pollution level of the target photovoltaic panel is ≥2, start the nozzle for water-assisted cleaning.

6. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 1, characterized in that: The method for adjusting the horizontality of the beam includes: the inclination sensor monitors the horizontality of the beam in real time and transmits data to the controller; the controller controls the extension and contraction of the hydraulic cylinder to adjust the horizontality of the beam according to the deviation value fed back by the inclination sensor.

7. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 4, characterized in that: The method for determining the cleaning frequency includes: Define the pollution increment function G1(dt)= F(x i ,y i ,z i ,t1+dt)- F(x i ,y i ,z i ,t1), and observe the change of G1(dt) by adjusting the cleaning interval dt: Initially set dt_min and dt_max as the preset minimum and maximum cleaning intervals; In the range of dt_min to dt_max, gradually adjust dt and calculate G1(dt); If G1(dt)∈(0,2], then the current dt is the optimal cleaning interval; If G1(dt) > 2, shorten dt and re-evaluate; If G1(dt) = 0, extend dt and re-evaluate; By iterating the above steps, the dt value that makes G1(dt) stable in the range of (0,2] is finally determined as the periodic cleaning frequency.

8. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 1, characterized in that: The robotic arm is a six-degree-of-freedom robotic arm.

9. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 1, characterized in that: The track surface is provided with anti-skid patterns.

10. The large-area cleaning robot for mountain / hilly photovoltaic power stations according to claim 1, characterized in that: The bottom of the gantry is equipped with a magnetic rail brake.

Images