NONLINEAR SIGNAL ANALYSIS OF DYNAMIC SYSTEMS BASED ON EMPIRICAL MODE DECOMPOSITION AND FAST FOURIER TRANSFORMATION METHOD FOR FAULT DETECTING
A. IRANMANESH *
Department of Civil Engineering, Vali-e-Asr University of Rafsanjan, Rafsanjan, Iran.
*Author to whom correspondence should be addressed.
Abstract
In this paper, a new method for detecting of faults in mechanical systems is introduced based on the conditional monitoring of vibrating systems. Hilbert- Huang Transform (HHT) is a powerful method for nonlinear and non-stationary vibrations analysis. This approach consists of two basic parts that are Empirical Mode Decomposition (EMD) and Hilbert Spectral Analysis (HSA). To obtain the reliable results and also accurate frequencies, Bedrosian and Nuttall theorems are satisfied to prevent the combination of phase and amplitude functions. Thus, different methods such as improved Hilbert-Huang transforms were proposed in the literature. According to the proposed method in this study, signals are decomposed to their Intrinsic Mode Functions (IMFs) and each IMF is investigated by Fast Fourier Transform (FFT). Using the suggested technique, which is entitled as EDM-FFT, the mechanical faults of an electro-motor are studied. It is seen that the classical FFT method is unable to detect all defects, while all of the characteristic frequencies are recognized using this method. Also, in spite of HHT scheme, utilizing the suggested method prevents the combination of phase and amplitude functions.
Keywords: Hilbert-Huang transformation, system fault, empirical mode decomposition, Fast Fourier transformation