Identification of electrical tree growth in insulation systems by fuzzy logic techniques based on partial discharge acquisition

Aim of this paper is to validate the effectiveness of an artificial intelligence technique devised for tree diagnosis in solid insulation systems. For this purpose, electrical and optical partial discharge (PD) measurements were carried out on specimens containing treed regions of different typologies ("bush" and "branch"). The results obtained by the above-mentioned diagnostic procedure were checked on the basis of the tree typology.