Comparison of EMD and EEMD in rolling bearing fault signal analysis
| Autor: | Ke Fang, Yimei Dai, Hongmei Qi, He-sheng Zhang |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Signal processing
business.industry 02 engineering and technology White noise Hilbert spectral analysis 021001 nanoscience & nanotechnology Fault (power engineering) Computer Science::Numerical Analysis 01 natural sciences Signal Hilbert–Huang transform Frequency domain 0103 physical sciences Medicine 0210 nano-technology business 010301 acoustics Algorithm Hilbert spectrum |
| Zdroj: | I2MTC |
| DOI: | 10.1109/i2mtc.2018.8409666 |
| Popis: | Classical Hilbert-Huang transform (HHT) is commonly used in bearing vibration signal analysis and fault feature extraction, which consists of two parts: Empirical Mode Decomposition (EMD) and Hilbert spectrum analysis. That mode mixing exists in EMD affects the results of Hilbert spectral analysis seriously. Aiming at the problem, the ensemble empirical mode decomposition (EEMD) is used to replace the EMD in the classical Hilbert Huang transform, and two methods are used to analyze the synthesized simulation signals respectively. The simulation results show that the EMD method decomposes the multi frequency signal into a series of Intrinsic Mode Function (IMF) effectively, but with the mode mixing phenomenon; EEMD can suppress mode mixing by introducing random white noise; for different fault signals, EEMD can still effectively extract bearing fault features when the EMD method fails. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|