Extended Abstract (344K)P2.5
The Empirical Mode Decomposition and Intermittent Turbulence
Christina Ho, Texas Tech Univ., Lubbock, TX; and X. Gilliam, J. A. Phillipson, and S. Basu
Characterizing and analyzing turbulence in a nocturnal stable boundary layer provides a great challenge in signal processing. In particular, the turbulence is weak, patchy, intermittent and highly non-stationary. For these reasons, traditional signal processing tools such as spectrum analysis often fail to catch such phenomena. In recent years, empirical mode decomposition (EMD) techniques have become a popular tool for nonlinear and non-stationary time series analysis. Using the EMD method, the non-stationary time series is decomposed into multiple intrinsic mode functions containing intrinsic time scales characteristics. In this paper, the EMD method is applied to data collected in a nocturnal stable boundary layer. Once the intrinsic modes have been separated from the time series, the important features such as the power and frequency of the particular scale is studied through the Hilbert transform. Finally, a statistical technique based on randomized distribution is then developed to detect the intermittent turbulence. In this test, the test statistic is based on the proper intrinsic mode functions and the critical value of the test is obtained from the incoherent exemplars through the randomization process of the time series.
Poster Session P2, Poster Session P2
Wednesday, 23 January 2008, 2:30 PM-4:00 PM, Exhibit Hall B
Previous paper Next paper