A New Algorithm to Identify Magnetizing Inrush Conditions Based on Instantaneous Frequency of Differential Power Signal

This paper presents a novel algorithm to recognize magnetizing inrush condition in power transformers by analyzing average differential power signal. The algorithm is independent of the value of transformer parameters and consumed power and relies only on the waveshape properties of the average power signal. These distinctive properties are obtained from the power signal by determining its instantaneous frequency. Extensive simulation studies presented in this paper demonstrate the merits of the proposed technique in discriminating between fault and inrush cases accurately in less than a quarter cycle. The accuracy and speed of the method is unaffected by changing parameters of the transformer or for different operating conditions of the power system, since it is based on a fundamental difference between iron-core saturation and other operating conditions. Since the employed signatures are extracted from the low-frequency spectrum of the power signal, the algorithm is robust against high-frequency noise.