A novel technique to discriminate inrush and fault in a single-phase transformer. (16th August 2010)

• Record Type: Journal Article
• Authors: Paraskar, S.R.; Beg, M.A.; Dhole, G.M.; Khedkar, M.K.
• Is Part Of: International journal of signal and imaging systems engineering. Volume 3:Number 1(2010)
• Page Start: 52
• Page End: 60
• Publication Date: 2010-08-16
• Subjects: ANNs -- artificial neural networks -- single-phase transformers -- fault detection -- DWT -- discrete wavelet transform -- magnetising inrush -- interturn faults -- transformer protection; Signal processing -- Periodicals; Imaging systems -- Periodicals; Information theory -- Periodicals; 621.3822
• Languages: English
• ISSNs: 1748-0698

Abstract

This paper presents an algorithm based on a combination of Discrete Wavelet Transforms (DWTs) and Feed-Forward Artificial Neural Network (FFANN) to discriminate magnetising inrush from interturn fault. Interturn faults are staged on custom-built transformer. DWT is used for feature extraction from the differential current during magnetising inrush and interturn faults and FFANN is used to discriminate magnetising inrush from interturn fault. An online algorithm is tested successfully on the custom-built transformer. It is found that the proposed method gives satisfactory results, and may be useful in the development of modern differential relay for transformer protection scheme.