基于视觉-光电传感器的光伏板清洁机器人研究

Research on Photovoltaic Plate Cleaning Robot Based on Vision Photoelectric Sensors

我国南方湿地及丘陵地区太阳能资源丰富,但路面条件差,光伏板安装位置高,导致灰尘和鸟粪等污染物难以清理,影响光伏电站的经济性和安全性.为解决地面及空间复杂的光伏电站的清洁难题,设计了一种基于视觉-光电传感器的光伏板清洁机器人,分析倾斜光伏板上机器人的附着条件和所需吸力.根据视觉传感器和光电传感器提供的信息,控制机器人自适应路径规划并执行清洁策略,实现光伏板表面灰尘的全面清洁和鸟粪的定位清洁.试验结果表明,光伏板清洁机器人的清洁效率为50 m^(2)/h,鸟粪的平均定位误差为1.38 mm,除尘率为91.16%,有利于促进太阳能光伏发电的提质增效.

Wetlands and hilly areas in southern China are rich in solar energy.However the poor roads and high locations of photovoltaic panels make it difficult to clean up pollutants such as the dust and the bird droppings,which affects the economy and safety of power plants.In order to solve the cleaning problem of the complex PV plants on the ground and in space,a photovoltaic panel cleaning robot based on the vision-photoelectric sensors have been designed,and the attachment conditions and the required suction of the robot on the tilted photovoltaic panels are analyzed.Based on the information provided by the visual and photoelectric sensors,the robot is controlled to adapt to the path planning and execute the cleaning strategies so that the comprehensive cleaning of dust on the surface of photovoltaic panels is achieved and the cleaning of bird droppings is positioned.The experimental results show that the cleaning robot has a cleaning efficiency of 50 m^(2)/h,an average localization error of 1.38 mm for the bird droppings and a dust removal rate of 91.16%,which is conducive to promoting the improvements of quality and efficiency in solar photovoltaic power generation.