Identification of Free Conducting Particles in Transformer Oils using PD Signals

Firuzi K. , Parvin V., Vakilian M.

2015 IEEE 11th International Conference on the Properties and Applications of Dielectric Materials (ICPADM), Sydney, Australia, 19 - 22 July 2015, pp.724-727 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/icpadm.2015.7295374
  • City: Sydney
  • Country: Australia
  • Page Numbers: pp.724-727
  • Keywords: Partial discharge, PCA Feature Extraction, SVM Classifier


Transformers are known as one of the most important equipment in power system transmission and distribution network. Safety of transformer insulation is determined mainly by its insulating oil dielectric strength. A major concern which threaten the withstand strength of a liquid insulation is the presence of particle contamination. One of the best methods to detect any abnormality and insulation weakness inside the transformer insulation is based on partial discharge (PD) measurement. Here, to identify the presence of conducting particles inside the transformer insulating oil, the general routine used for PD recognition is employed. This process involves the following steps: current signal measurement, PD pulse capture, PD signal parameters extraction, signal categorization based on existing data bases, and finally its diagnosis and finding the source of PD. The PD pulse characteristics related to a floating conductive particle, with different shapes and sizes, under quasiuniform electric field is studied in this paper. These characteristics related to the particles in transformer oil, besides data mining techniques are employed to analyze the recorded PD measured signals in time domain and subsequently to specify the shape and size of particle. PCA feature extraction method is applied on the frequency domain data, then SVM classifier is used to classify the recorded data. Results based on experimental training and testing data indicate that this method using PD signal information provides a 97% classification success rate.