•  
  •  
 

Keywords

zero-value insulator; quadcopter; spatial electric field; charged detection; BP neural network

Abstract

Insulators play the role of ensuring the insulation of conductors to the ground in the power system, which is an important piece of equipment for the external insulation of transmission and distribution, as well as a reliable guarantee for the safety and stability of the power system. Zero-value faults in insulators can cause serious problems in the operation of the power system. By detecting the local electric field by quadcopter, it is possible to determine whether the insulator is a zero-value unit. In order to figure out the reasonable detection distance and propose a rapid identification criterion, an insulator detection and device structure method based on a quadcopter is proposed, and the equivalent model of a 500 kV insulator string and quadcopter is designed. The detection results of the quadcopter at different detection distances are simulated and analyzed, so as to delineate a reasonable detection distance range. By analyzing the detected electric fields in different directions, an optimal detection method is proposed. By combining with the back propagation (BP) neural network algorithm, the measured discrete data are accurately fitted to the radial electric field distribution plot. Research results show that when the detection distance is 70 ~ 200 mm, and the minimum absolute value of the slope of the image curve is less than 0.015, the insulator can be determined as a zero-value unit. When the detection distance is less than 70 mm, and the variance of the electric field amplitude of the image is less than 55, the insulator is determined as a zero-value unit. The criterion accuracy of the method is more than 98.5%, and this achievement can provide technical support for the operation and maintenance of external insulation of transmission and distribution.

DOI

10.19781/j.issn.1673-9140.2026.01.026

First Page

277

Last Page

288

Download

Share

COinS