Abstract
In this work, the loss of insulation life such as affecting a power transformer operating under harmonic distortion is predicted or forecasted using the following computational techniques: artificial neural network, fuzzy logic and adaptive neuro-fuzzy inference system. The IEEE C57.91, IEEE C57.110 as well as the IEC 60076-7 loading guides regulations are used to estimate the transformer insulation loss of life. The transformer used in this work consisted of an oil type delta-wye step down with a rating of 1 MVA 11/0.4 kV located in a hospital facility's medium-voltage distribution substation. Previous studies that have been conducted on transformer insulation loss of life focused on an analysis based only on a 24-hour loading. This analysis is based on a medium-term period of 31 days. Since environmental temperature and load fluctuations with time affect daily insulation degradation rate, medium or long-term analysis will produce a better understanding of how the insulation of a distribution transformer degrades...
M.Phil. (Electrical and Electronic Engineering)