Detection of Transients in Vibration Signals Using Reverse Dispersion Entropy

Abstract

In this work, the reverse dispersion entropy (RDE) is used to process the squared envelope (SE) signal in order to detect nonstationarites. Based on the idea of spectral kurtosis (SK) and kurtogram, the squared envelope signal is first extracted by applying the short time Fourier transform to vibration signal. Then, as an alternative to negative Shannon entropy, the RDE is used to process the squared envelope to detect the range of frequencies at which the transients occur. The RDEgram color-coded map is used to represent the RDE values as a function of frequency and frequency resolution from which the ideal filter parameters can be inferred. Once the best frequency and frequency bandwidth pair are found, an optimum finite impulse response filter can be designed to filter the original vibration signal. The proposed method is tested against simulated and actual vibration signals and proved to be superior to existing methods.