Second order multi-synchrosqueezing transform for rub-impact detection of rotor systems

Abstract

Rub-impact fault is one of the most common and severe failure modes for rotating machinery. Disclosing the changing dynamic in rub-impact fault signal could provide a basic foundation for rub-impact detection of rotor systems. In this study, a dynamic model of a flexible rotor system is established to simulate the rub-impact fault signal to help us understand the dynamic characteristics of rub-impact fault signal. Meanwhile, considering that an advanced time-frequency analysis method is of vital importance in an effective rub-impact fault detection method. A new time-frequency analysis algorithm termed as second order multi-synchrosqueezing transform is also proposed in this study, in which a second order two-dimensional instantaneous frequency estimation is embedded into a multi-synchrosqueezing framework. It is shown that the proposed method could produce a time-frequency representation result with a high time-frequency resolution. Assisted by the proposed method, the impulse shaped amplitude modulated phenomenon in high frequency band and the fast varying frequency modulated phenomenon around the rotation frequency and its fractional or super harmonics are clearly revealed through the analysis of rub-impact fault signals from both dynamical simulation and experimental cases.