CT Saturation Detection Based on Waveshape Properties of Current Difference Functions

Abstract

Derivative-based approaches have been widely used for current-transformer (CT) saturation detection. Previous approaches dealt with the magnitude of current derivatives. This paper, however, proposes a novel CT saturation detection algorithm that is subject to the waveshape properties of current derivatives. Two indices are introduced to extract certain key features of a current waveshape using its first two difference functions. Combining these two indices, CT saturation is rapidly and reliably detected. This method operates correctly irrespective of CT parameters, fault current characteristics, and power system conditions. Derivative-based techniques are suspected to be noise sensitive. Meanwhile, appropriate measures are taken, and the robustness of this method against noise is guaranteed. Extensive analysis of various real and simulated currents confirms that this algorithm can meet the rigorous speed and accuracy standards of industrial relays.