Time-Frequency Demodulation Analysis Based on LMD and Its Applications

Abstract

A time-frequency demodulation technique based on local mean decomposition (LMD) is proposed for rotating machine diagnosis. In addition, methods for boundary processing and for determining the step size of the moving average are presented to improve LMD algorithm. Instantaneous amplitude (IA) and instantaneous frequency (IF) of the signal can be achieved using the improved LMD method. A well-constructed description of the derived IAs and IFs is represented in the form of instantaneous time-frequency spectrum (ITFS), which preserves both the time and frequency information simultaneously. Results of three synthetic signals indicate that the proposed method is much better in extracting the comprehensive carrier and modulated components, compared with Hilbert-Huang transform and stationary wavelet transform. The validity of the technique is further demonstrated on the rotor system and a gearbox. The transient fluctuations of the IF and the impulsive signatures can be successfully identified in the ITFS. Moreover, it has been demonstrated that the proposed time-frequency demodulation technique is much more effective and sensitive than the other methods in detecting impulsive and modulated components.