Title :
A De-Noising Method for Velocity Signal in Impinging Stream Mixer Based on Empirical Mode Decomposition
Author :
Zhang, Jianwei ; Miao, Chao
Author_Institution :
Sch. of Mech. Eng., Shenyang Inst. of Chem. Technol., Shenyang, China
Abstract :
The velocity time series signal of the flow field in impinging stream mixer (ISM) contains a lot of dynamic information. It is the comprehensive reflection of many factors such as the physical properties, the motion characteristics and the flow structures of the fluid. The measured velocity signal denoising process is the key to improve the signal reliability and accuracy. A de-noising method for velocity signal in ISM based on empirical mode decomposition (EMD) was proposed. To distinguish noise and useful signal in intrinsic mode function (IMF), the consecutive mean square error (CMSE) was used, then the modes reflecting the important structures of the signal were combined together to form partially reconstructed denoised signal. The results indicated that this method can efficiently and adaptively remove noise, and this method can not be affected by subjective parameters.
Keywords :
mean square error methods; reliability; signal denoising; time series; consecutive mean square error; empirical mode decomposition; flow field; impinging stream mixer; intrinsic mode function; motion characteristics; signal reliability; velocity signal denoising method; velocity time series signal; Chemical technology; Fluid dynamics; Mechanical engineering; Nanoscale devices; Noise reduction; Reflection; Signal denoising; Signal processing; Wavelet analysis; Wavelet coefficients;
Conference_Titel :
Image and Signal Processing, 2009. CISP '09. 2nd International Congress on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-4129-7
Electronic_ISBN :
978-1-4244-4131-0
DOI :
10.1109/CISP.2009.5301049
