| As a major power user,China is in a leading position in the world in terms of power system related industry technology.At present,the inspection of power panel cabinets in substations is still done manually.Manual inspection has uncertain factors such as personal’s professional quality,technical experience,and work enthusiasm,while robot inspection can make up for the shortcomings of manual inspection in these aspects,and improve the efficiency and quality of operation and maintenance.At the same time,it can have both safety and economy.Especially after this Covid-19,the realization of remote intelligent supervision of power screen cabinets has become an urgent need of the power industry.In this paper,an autonomous patrol robot based on image recognition is used to replace the staff to inspect the power protection panel of the substations.The robot uses multi-sensor fusion technology to quickly build environmental map;uses algorithms such as adaptive Monte Carlo particle filtering achieve autonomous obstacle avoidance,precise poisoning and path planning of the robot,so as to ensure that the robot can navigate quickly,accurately and stably in front of the equipment,in order to carry out follow-up inspection work.The robot’s inspection method takes into account the problems of the large number of switches,different colors,small number fonts,complex lighting conditions,etc.Combining image recognition and deep learning algorithms,we have achieved accurate identification of the pressure plate of the power protection screen cabinet.It avoids the fatigue caused by long-term repetitive work in the manual verification process,which may bring about missed and false detections,and ensures the safe operation of the power grid system and the personal safety of the inspectors.At the same time,it realizes the real-time monitoring and intelligent information management of the power system.The inspection robot for substation power screen cabinets developed in this paper adopts robotics technology,internet and internet of things technology to build a complete monitoring platform.It can gradually and effectively replace manual inspections without changing the working methods of many devices that have been put into use.To build an unattended remote intelligent supervision system and improve the advanced nature of grid monitoring and management,this will play a leading role in the power industry and provide a demonstration for industrialization. |
