Real-time implementation of the digital Taylor-Fourier transform for identifying low frequency oscillations

This paper proposes the use of the digital Taylor-Fourier transform (DTFT) for identifying low frequency oscillations in power systems. The implementation has been performed on the CompactRIO (cRIO) platform and its associated libraries, using a computational efficient DTFT calculation. The platform generates the signal spectral decomposition, yielding mono-component signals extracted by a filter bank. The spectral analysis is accomplished by means of a sliding-window that advances each new sample, providing reconstructed signals, their amplitude estimates, and information of their instantaneous damping and frequency. The identification process is applied on both simulated and actual signals. Experimental results confirm the proposition's performance, precision, and reliability.