Research On The Motion Control System Of Substation Water Washing Robot Based On Visual Positioning

With the continuous development of society,people’s demand for stable power supply is growing,and the steady operation of power supply system is the basis for ensuring continuous and stable power supply.Therefore,regular maintenance of substation transmission and substation equipment can ensure the regular functioning of power equipment.As one of the important transmission and substation equipment,insulators are exposed to the outdoor environment for a long time.Due to the strong electric field and the harsh natural environment,a large amount of pollutants accumulate on the surface of insulators to form dirt,which can easily cause flicker and lead to safety accidents and prevent the normal operation of the power supply system,so it is necessary to treat the pollutants on the surface of insulators on a regular basis.At present,insulators are mainly cleaned by hand on a regular basis,which has the disadvantages of low automation level,high labor intensity and high danger.In view of the drawbacks of the traditional insulator cleaning method,this thesis is based on the research of the water rinsing robot for substation strip operation,and carries out the research on the key tasks such as insulator identification and positioning algorithm and robot motion control in substation,and the main research and work are as follows:(1)According to the operating environment of the substation water washing robot,combined with the analysis of the functional requirements of the robot,the main structural parameters of the robot were designed,the overall structural composition of the water washing robot was introduced,and the structural design of the robot’s mobile chassis platform and the five-degree-of-freedom robot arm was carried out,the total architecture of the motion control system of the water washing robot was determined,and the selection of the hardware for the motion control system was accomplished.(2)The substation insulator target detection method is studied,and the current mainstream target detection algorithms are introduced.To achieve accurate and efficient insulator identification in substations,a substation insulator identification approach based on improved YOLOv5 is introduced,which uses the Mosaic-9 data enhancement method to effectively improve the generalization ability of the network model and introduces the attention module in the backbone network.enhance the attention of the network model to insulators and improve the target detection accuracy,while the Ghost module is used to replace the C3 module with partial convolutional structure in Neck network to ensure the real-time target detection.Through experimental comparison and analysis,the volume of the improved YOLOv5 network model is reduced by 21.5%,the average detection speed is77.2f/s,and the average detection accuracy reaches 90.8%,which verifies the effectiveness of the method proposed in this thesis.(3)Based on the insulator target detection,the insulator localization method based on binocular vision technology is proposed.The method first uses camera calibration to obtain the internal and external parameters of the camera,and uses image correction to enhance the subsequent stereo matching effect,and finally uses the SGBM algorithm to find the corresponding matching points of the target point in the two images,so as to calculate the spatial coordinates of the target insulator by triangulation.The insulator positioning method based on binocular vision technology has a maximum absolute error of 16.7 cm and an average absolute error of 6.27 cm at a detection distance of 6 m.The positioning accuracy can meet the requirements of the substation water washing robot.(4)The research on the motion control method of water washing robot is carried out,firstly,the kinematic model of water washing robot is established,the positive and negative kinematics of the robot is analyzed,and the spatial accessibility of the robot is verified by simulating the robot operation space using Monte Carlo method,secondly,the hand-eye relationship of the robot is analyzed,and a double closed-loop visual servo system is formed,and a motion control strategy combining the position-based visual control A motion control strategy combining position-based visual control method and fuzzy adaptive PID control algorithm is proposed.The simulation results show that the proposed motion control strategy can effectively complete the automatic targeting and flushing of insulators with good control effect.