Design of Deep Learning Algorithm in the Control System of Intelligent Inspection Robot of Substation

Abstract

With the development of the scale of the power system and the maturity of robotics, the use of inspection robots instead of manual inspection can effectively improve the efficiency of inspection and realize the intelligent management of the power grid. In this paper, a control system for substation inspection robot is designed and implemented for the substation inspection environment. First, a convolutional neural network algorithm under deep learning (DL) is proposed to extract the characteristics of the inspection robot, and then related technologies for inspection robot control are proposed, including navigation and positioning, motion control, power supply management, safety and anti-collision, and finally, it can be obtained through the test of the control system. The inspection robot has good tracking capabilities. In addition, the real trajectory is basically consistent with the set trajectory, which also shows the robustness of the system. Therefore, the use of DL to design the control system of the intelligent inspection robot of the substation is of great research value. In this paper, it is hoped that the deep learning-based convolutional neural network algorithm can be used to extract the features of inspection robots, which can effectively prove that the robots have good tracking ability and promote the power grid inspection to a certain extent. This paper provides reference value for realizing the intelligentization of power grid selection.