Power transformer paper insulation assessment based on oil measurement data using SVM-classifier

Abstract

Oil immersed paper insulation condition is a crucial aspect of power transformer’s life condition diagnostics. The measurements testing database collected over the years made it possible for researchers to implement classification analysis to in-service power transformer. In order to generate a reliable model, more studies related to machine learning implementation to power transformer assessment need to be done. In this article, the objective of the study is to develop reliable new approach in transformer oil-immersed paper insulation condition assessment based on SVM-classifier model using its oil measurements. The measurements data (dielectric characteristics, dissolved gas analysis, and furanic compounds) of 149 transformers with primary voltage of 150 kV had been gathered and analyzed. The algorithm employed for developing classification model is Support Vector Machine (SVM). The model had been trained and tested using different datasets. Several different models have been created and the best has been chosen, resulting in 90.63% accuracy in predicting the oil-immersed paper insulation condition. Further implementation was executed to classify oil-paper condition of 19 transformers which Furan data is not available. The classification results were combined, reviewed, and compared to conventional assessment methods and standards. The comparation confirmed that the model developed has the ability to do classification of current oil-paper condition for the transformer population observed, based on Dissolved Gasses and Dielectric Characteristics.