A FUZZY INFORMATION APPROACH TO INTEGRATING DIFFERENT TRANSFORMER DIAGNOSTIC METHODS

Methods to identify transformer fault conditions before they deteriorate to a severe state include dissolved gas analysis, liquid chromatography, acoustic analysis, and transfer function techniques. All of these methods require some experience in order to correctly interpret observations. Researchers have applied artificial intelligence concepts in order to encode these diagnostic techniques. These attempts have concentrated on only a single technique and have failed to fully manage the inherent uncertainty in the various methods. In this paper, a theoretic fuzzy information model is introduced. An inference scheme which yields the ''most'' consistent conclusion is proposed. A framework is established that allows various diagnostic methods to be combined in a systematic way. Numerical examples demonstrate the developed system.