Wavelet-Based Scheme for Detection of Torsional Oscillation

A new scheme for torsional oscillation detection is proposed. The scheme is based on the biorthogonal wavelet transform (WT) and complex WT. The first step of the method is to design a biorthogonal low-pass filter, which filters the high order harmonics. A complex wavelet is then used to extract the oscillation feature from the remnant signal. The effectiveness of the proposed scheme is verified with mathematical and simulation models. Furthermore, as the Mallat algorithm and the recursive computation method are used for the proposed scheme, it is easy to be implemented for real-time applications.