Detection of signal component modulations using modulation intensity distribution

Modulations in vibration signals that are generated by rotating machinery carry a lot of useful information about health condition. This paper proposes a technique for detection of modulations, called modulation intensity distribution (MID), which may be considered as a generalization of the spectral correlation density. It applies to most common types of modulations in mechanical systems thanks to an open module that can be tailored accordingly by the user. This leads to specific versions of the MID based on the RMS value, the spectral correlation or the kurtosis, which apply to the detection of amplitude modulations of either discrete or random carriers. The algorithm for calculation of the MID is described and tested on both simulated signal and experimental vibrations. The utility of MID in the diagnostic of rolling element bearing is demonstrated to be a natural extension of narrow band envelope analysis.