Outlier Detection of the Power Transformer DGA Fault Data Based on Ensemble Model

Abstract

In this paper., a new outlier detection method is proposed for the validity of DGA data for online monitoring of power transformers. The method aims to evaluate the validity of the data remitted to the fault database and uses a weighted ensembling of three outlier detection algorithms with different principles in order to avoid the uncertainty of a single model to reject outliers. The experimental results show that the proposed method has better performance in handling outliers in DGA fault data.