Random distribution and adsorption characteristics of metal particles and visualization algorithm for DC GIS/GIL

The metal dust moving inside the HVDC GIS/GIL will be easily adsorbed to the surface of the insulator and cause surface discharge. The study of the movement and adsorption characteristics of metal dust is extremely important. In this paper, NPLE image processing method is proposed that uses unified indicators to observe and count metal dust. The movement rules of metal dust particle groups inside DC GIS/GIL and the adsorption characteristics on the insulator surface are studied. The effect of metal particle traps on metal dust capture efficiency is explored. The results show that the range of concentrated distribution of metal dust inside DC GIS/GIL is basically stable, but there is a certain degree of randomness. The concentrated distribution range of metal dust is 148.1°, and the range where a large amount of metal dust is adsorbed on the surface of the basin insulator is 151.5°. The two ranges are very close. Assembling a particle trap can reduce the amount of dust adsorption by up to 99.86% in this experiment. Further simulation analysis found that areas with strong normal field strength on the surface of the insulator are prone to absorbing metal particles. The results can provide a basis for suppression and capture strategies of metal particles inside DC GIS/GIL.