A Method of Prediction for Transformer Malfunction Based on Oil Chromatography

Abstract

Electric power transformer is one of the most necessary part in power system, and hence, it’s significant to diagnose the transformer malfunction in advance; A methods of prediction for transformer malfunction based on oil chromatography is described; 4 models for time series prediction are illustrated, and the specific methods for model identification and ordering are explained; The time series model was applied to predict transformer malfunction in the oil chromatography analysis example, and accurate results were obtained, which shows that the method described in this paper can effectively predict the concentration of dissolved gas in transformer oil in future, and diagnose the types of malfunctions so that meet the actual need of projects.