Aiming at the problem that insulator image backgrounds are complex and fault types are diverse, which makes it difficult for existing deep learning algorithms to achieve accurate insulator fault diagnosis, an insulator fault diagnosis method based on multi-mechanism optimization YOLOv8-DCP is proposed. Firstly, a feature extraction and fusion module, named CW-DRB, was designed. This module enhances the C2f structure of YOLOv8 by incorporating the dilation-wise residual module and the dilated re-param module. The introduction of this module improves YOLOv8’s capability for multi-scale feature extraction and multi-level feature fusion. Secondly, the CARAFE module, which is feature content-aware, was introduced to replace the up-sampling layer in YOLOv8n, thereby enhancing the model’s feature map reconstruction ability. Finally, an additional small-object detection layer was added to improve the detection accuracy of small defects. Simulation results indicate that YOLOv8-DCP achieves an accuracy of 97.7% and an mAP@0.5 of 93.9%. Compared to YOLOv5, YOLOv7, and YOLOv8n, the accuracy improved by 1.5%, 4.3%, and 4.8%, while the mAP@0.5 increased by 3.0%, 4.3%, and 3.1%. This results in a significant enhancement in the accuracy of insulator fault diagnosis.

To ensure the safe operation of high-voltage transmission line inspection robots during downhill descents without power and extend their range after a single charge, this paper proposes an energy-saving speed control method for the inspection robot’s walking wheel motor on downhill slopes by integrating feedback braking and fuzzy PID control. By combining the parameter equation of the overhead catenary line and the structural characteristics of the overhead transmission line, this paper analyzes the relationship between the driving torque of the inspection robot’s wheels and the horizontal displacement along the transmission ground wire before and after descending. Based on this analysis, a speed control and energy recovery scheme is developed for the inspection robot, which combines front-wheel feedback braking with rear-wheel regenerative braking. The fuzzy PID method is utilized to adjust the PWM duty cycle to achieve energy-efficient speed control of the inspection robot’s rear walking wheels. Additionally, to improve the energy density and specific power of the robot’s energy storage unit, a composite power source consisting of lithium batteries and supercapacitors is employed to recover energy from the front walking wheels through feedback braking. The combined simulation results indicate that, compared to fuzzy control and PID control, fuzzy PID control better regulates the robot’s speed under varying slopes, wind resistance, and cable roughness. A downhill speed control system for the inspection of the robot’s walking wheel motor was designed, and its effectiveness was validated through simulated high-voltage line experiments. The fuzzy PID control was demonstrated to effectively maintain the rear walking wheel speed within the targeted range during downhill descents. When descending along a fixed 30° angle cable, the fuzzy PID control resulted in an increase of 5.28% and 14.26% in the state of charge (SOC) of the supercapacitor compared to PID control and fuzzy control, respectively. Moreover, when descending along fixed angle cables of 10°, 20°, and 30°, as well as a variable angle cable ranging from 30° to 0°, the SOC of the supercapacitor increased by 17.55%, 26.25%, 38.45%, and 31.29%, respectively. This demonstrates the effective absorption of regenerative braking energy during the robot’s downhill movement.