Finite-Element Performance Evaluation of On-Line Transformer Internal Fault Detection Based on Instantaneous Voltage and Current Measurements

Abstract

This paper investigates the performance of a recently proposed online transformer internal fault detection technique through detailed non-linear three-dimensional finite element modelling of the windings, magnetic core and transformer tank. The online technique considers correlation of transformer instantaneous input and output voltage difference and input current at the power frequency and uses the ellipse shape ΔV-I locus as the finger print of the transformer that could be measured every cycle to identify any incipient faults. The technique is simple, fast and suitable for online monitoring of in-service transformers. A detailed three-dimensional finite element model of single-phase transformer is developed and various physical winding deformations with different fault levels are simulated to assess their impacts on the online ΔV-I locus. As transformer field testing under various internal fault conditions cannot be easily conducted, the main contributions of this paper are accurate finite element based implementation, testing and performance evaluation of the online fault detection approach. Furthermore, the impact of winding short circuit fault on the ΔV-I locus has been also measured and validated.