500kV变电站接地作业机器人的作业控制方法研究OA

To address the inefficiency and high risk of manual grounding operations in 500 kV substations,a com-pact and lightweight grounding robot using drone rope hanging and robot suspension system is developed.The YOLOv11 target detection algorithm is employed to recognize the monkey head clamp and conductor,and bin-ocular cameras are used to obtain the three-dimensional coordinates of the relevant components.An MSFNet-based gap recognition algorithm between the monkey head clamp and conductor is proposed to assist the robot in controlling the gap during the clamping operation.A three-ring control based on position,speed,and current is used to hang the monkey head clamp on the conductor,followed by RBF-PID control of the rotational manipu-lator to secure the clamp on the conductor.Simulations and field tests show that the visual recognition algorithm achieves overall average recognition rates of 97%and 87.5%for the monkey head clamp and conductor,respec-tively.The three-dimensional coordinates measured by binocular camera deviate less from actual coordinates by 1 cm,meeting operational requirements.The RBF-PID control,aided by feedback from the gap recognition al-gorithm,effectively and robustly completed the clamping operation,confirming the control method's feasibili-ty.The designed grounding robot can efficiently and high-quality perform grounding operations in substations,and have significant practical application value and significance.