Variants of Hilbert–Huang Transform with Applications to Power Systems’ Oscillatory Dynamics

Power system dynamic processes may exhibit highly complex spatial and temporal dynamics and take place over a great range of timescales. When frequency analysis requires the separation of a signal into its essential components, resolution becomes an important issue. The Hilbert–Huang transform (HHT) introduced by Huang is a powerful data-driven, adaptive technique for analyzing data from nonlinear and nonstationary processes. The core to this development is the empirical mode decomposition (EMD) that separates a signal into a series of amplitude- and frequency-modulated signal components from which temporal modal properties can be derived. Previous analytical works have shown that several problems may prevent the effective use of EMD on various types of signals especially those exhibiting closely spaced frequency components and mode mixing. The method allows a precise characterization of temporal modal frequency and damping behavior and enables a better interpretation of nonlinear and nonstationary phenomena in physical terms. This chapter investigates several extension to the HHT. A critical review of existing approaches to HHT is first presented. Then, a refined masking signal EMD method is introduced that overcomes some of the limitations of the existing approaches to isolate and extract modal components. Techniques to compute a local Hilbert transformation are discussed and a number of numerical issues are discussed. As case studies, the applications of the various EDM algorithms in power system’ signal analysis are presented. The focus of the case studies is to accurately characterize composite system oscillation in a wide-area power network.