TY - GEN

T1 - Features of the Hilbert-Huang transformation and examples of its application for analysis of nonstationary processes in flows

AU - Mironov, D. S.

AU - Lebiga, V. A.

AU - Zinovyev, V. N.

AU - Pak, A. Yu

AU - Miau, J. J.

AU - Chung, K. M.

N1 - Publisher Copyright: © 2018 Author(s).

PY - 2018/11/2

Y1 - 2018/11/2

N2 - One of the main tasks of experimental gas dynamics is to study the mechanisms of formation and evolution of fluctuations of flow parameters that arise when flowing around the studied bodies. The processing of the experimental data in this case is reduced to the extraction of the component from total signal due to the influence of the process under study. As a rule, pulsation processes are of non-stationary nature, and for the most effective analysis of such phenomena it is necessary to use special methods of time-frequency analysis. One such method proposed by N. Huang et al. is Hilbert-Huang transformation. This transformation recently finds more and more applications for the analysis of various processes. Hilbert-Huang transform consists of two main stages: an empirical mode decomposition and Hilbert transform. As a result of the decomposition, the signal is represented as a finite sum of so-called intrinsic mode functions. These modes are essentially a basis and unique for each signal. To each of the intrinsic modes, the Hilbert transform can be applied, with the result of which any can compute amplitudes and frequencies of these modes for each instant of time. Possibilities and features of empirical mode decomposition procedure and of the overall method are demonstrated. As an examples results of decomposition of the experimental signal of a thermoanemometer are considered. These signals are obtained in various flows characterized by a high level and a significant instability of flow fluctuations.

AB - One of the main tasks of experimental gas dynamics is to study the mechanisms of formation and evolution of fluctuations of flow parameters that arise when flowing around the studied bodies. The processing of the experimental data in this case is reduced to the extraction of the component from total signal due to the influence of the process under study. As a rule, pulsation processes are of non-stationary nature, and for the most effective analysis of such phenomena it is necessary to use special methods of time-frequency analysis. One such method proposed by N. Huang et al. is Hilbert-Huang transformation. This transformation recently finds more and more applications for the analysis of various processes. Hilbert-Huang transform consists of two main stages: an empirical mode decomposition and Hilbert transform. As a result of the decomposition, the signal is represented as a finite sum of so-called intrinsic mode functions. These modes are essentially a basis and unique for each signal. To each of the intrinsic modes, the Hilbert transform can be applied, with the result of which any can compute amplitudes and frequencies of these modes for each instant of time. Possibilities and features of empirical mode decomposition procedure and of the overall method are demonstrated. As an examples results of decomposition of the experimental signal of a thermoanemometer are considered. These signals are obtained in various flows characterized by a high level and a significant instability of flow fluctuations.

UR - http://www.scopus.com/inward/record.url?scp=85056326638&partnerID=8YFLogxK

U2 - 10.1063/1.5065198

DO - 10.1063/1.5065198

M3 - Conference contribution

AN - SCOPUS:85056326638

VL - 2027

T3 - AIP Conference Proceedings

BT - 19th International Conference on the Methods of Aerophysical Research, ICMAR 2018

A2 - Fomin, null

PB - American Institute of Physics Inc.

T2 - 19th International Conference on the Methods of Aerophysical Research, ICMAR 2018

Y2 - 13 August 2018 through 19 August 2018

ER -