An effective filtering algorithm to mitigate transient decaying DC offset

Abstract

Phasor estimation as performed by the discrete Fourier transform (DFT) is adversely affected by the decaying dc offset which occurs following a fault. This paper presents a new algorithm based on the Fourier transform, which effectively mitigates the decaying dc offset. The proposed method uses the even and odd samples of the estimated amplitude. The method is based on the fact that, in phasor domain, the even and odd samples in a purely sinusoidal signal are almost the same, whereas in an exponential signal, especially at the beginning of the fault which has higher slope, they are different. It has been shown that subtracting the even samples from the odd samples in the estimated amplitude of the original signal effectively reduces the decaying dc offset effect and it gives the amplitude of the fundamental frequency with high accuracy. The proposed algorithm is able to mitigate any level of decaying dc offset in one cycle and there is no need to any more time for its calculation. Also, unlike other methods it does not extract the decaying dc parameters that cause delay in convergence to the final value. To demonstrate the effectiveness of the proposed algorithm it is compared with the conventional full cycle DFT method. The results show that the proposed algorithm is much faster than the conventional DFT method over a wide range of decaying dc offsets.