Hht Analysis And Application Of Mechanical Vibration Signals

As a new technology of time-varying non-stationary signal analysis, Hilbert-Huang Transform (HHT) is one of the popular research in the field of modern signal processing. It includes Empirical Mode Decomposition (EMD) and Hilbert transform two parts, in which EMD is the core part. Any complicated signal can be decomposed into a finite number of Intrinsic Mode Function (IMF) by EMD, and then using Hilbert transform to these IMF components can obtain instantaneous frequency as well as related concepts. With the advancement of technology and social development, HHT technology is increasingly applied to radiated noise of moving targets underwater, blasting vibration, earthquake, medical signals and mechanical equipment fault diagnosis and other aspects, which is possessed of great research value. Main contents of this paper include the following aspects:First of all, the paper expatiates on the basic theory and related concepts about HHT, analyses EMD decomposition algorithms and meaning in physics of HHT time-frequency spectrum, HHT energy spectrum as well as HHT marginal spectrum. Based on these, two main factors which influenced on the accuracy and speed of EMD-phenomenon of mode fission in noise signal analysis by EMD and end effect have been put forward.Secondly, compared with the traditional Fourier de-noising method, wavelet de-noising method, and EMD de-noising method accompanied emergence of HHT, this paper presents a new EMD combined with wavelet de-noising method, to apply to the background noise denoising of bearing signal. Through selecting white noise and Rayleigh distributed noise as the background noise signal, the indicator of signal to noise ratio and correlation coefficient show that the method has good de-noising effect.Again, the paper introduces a variety of classical power spectrum estimation methods, and for the conventional power spectrum estimation analysis and extraction method of signal line spectrum. In the paper, a new classical power spectrum estimation combined with EMD method has been put forward to analyze and extract the line spectrum, and makes simulation signals and the actual datas about bearing processing, thus demonstrating that the method has a better frequency resolution, at the same time by setting the characteristic quantities to carry out bearing fault detection.Finally, from both time and frequency two aspects, the paper studys the capabilities of HHT spectrum. Compared to FFT spectrum, STFT time-frequency spectrum, Wigner-Ville distribution and morlet wavelet spectrum, theoretic analysis and simulation experiment results show that HHT spectrum has perfect time-frequency concentration, eliminates the cross-time interference and good time-frequency resolution characteristics. It is confirmed by appling to mutations in the signal once again, so this paper recommends that the method is also applicable to fault signal detection.