Discrimination between magnetizing inrush and Interturn fault current in transformer: Hilbert transform-ANN approach

摘要

Transformer is One of the most important and expensive components in the power system. A wide variety of abnormal conditions and faults are occurred in it. The most difficult types of faults to detect within the transformer are Interturn faults. Hence it needs to be protected and requires special maintenance due to its importance and cost. Though there are various methods available for health monitoring of transformer, a focused technique is necessary for protection purpose. This paper presents a new algorithm for discrimination between magnetizing inrush and Interturn fault current using Hilbert transform and Artificial Neural Network. The performance of this algorithm is demonstrated on custom built mains feed three phase transformer in the laboratory. It is found that the proposed algorithm gives satisfactory results, and with the help of this algorithm, modern differential relay's development for transformer protection scheme can be done.