Abstract
The Hilbert-Huang transform (HHT) is a relatively new signal processing technique that offers several advantages over traditional Fourier transform analysis. The HHT empirically decomposes a signal into multiple components called intrinsic mode functions (IMF), then returns the instantaneous frequency and energy for each IMF. Having each IMF’s instantaneous frequency and energy for every data sample offers greater frequency resolution in both 2-D and 3-D spectral analysis—a clear advantage when compared to the Fourier transform’s well-known time-frequency resolution challenge. Because the HHT performs signal decomposition empirically no model or prior knowledge about the system is needed, making the HHT a useful technique for analyzing and giving physical interpretation to nonlinear, non-stationary systems. The Fourier transform, however, inherently only provides physical interpretation to linear, stationary systems; increasing the required expertise for use on nonlinear, non-stationary systems. This presentation will give an overview on the HHT algorithm, common modifications, advantages and disadvantages, how its results compare to the Fourier transform, and examples of successfully using the HHT on vibro-acoustic data for monitoring machine wear.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call