The application of Bayesian network theory in transformer condition assessment

In order to assess transformer´s status and forecast its potential fault, this paper applied Bayesian network classifier into transformer fault diagnosis, combined with dissolved gas analysis and other electrical test results, and thereby created a transformer fault synthetic diagnosis method. Build up transformer´s fault diagnosis model based on Naive Bayesian Classifier and Tree Augmented Naive Bayesian Classifier respectively, and verify their validity by instance.