Zhiming Lu

An amplitude-frequency study of turbulent scaling intermittency using Empirical Mode Decomposition and Hilbert Spectral Analysis

Hilbert-Huang transform is a method that has been introduced recently to decompose nonlinear, nonstationary time series into a sum of different modes, each one having a characteristic frequency. Here we show the first successful application of this approach to homogeneous turbulence time series. We associate each mode to dissipation, inertial range and integral scales. We then generalize this approach in order to characterize the scaling intermittency of turbulence in the inertial range, in an amplitude-frequency space. The new method is first validated using fractional Brownian motion simulations. We then obtain a 2D amplitude-frequency representation of the pdf of turbulent fluctuations with a scaling trend, and we show how multifractal exponents can be retrieved using this approach. We also find that the log-Poisson distribution fits the velocity amplitude pdf better than the lognormal distribution.

Arbitrary-order Hilbert spectral analysis for time series possessing scaling statistics: Comparison study with detrended fluctuation analysis and wavelet leaders

In this paper we present an extended version of Hilbert-Huang transform, namely arbitrary-order Hilbert spectral analysis, to characterize the scale-invariant properties of a time series directly in an amplitude-frequency space. We first show numerically that due to a nonlinear distortion, traditional methods require high-order harmonic components to represent nonlinear processes, except for the Hilbert-based method. This will lead to an artificial energy flux from the low-frequency (large scale) to the high-frequency (small scale) part. Thus the power law, if it exists, is contaminated. We then compare the Hilbert method with structure functions (SF), detrended fluctuation analysis (DFA), and wavelet leader (WL) by analyzing fractional Brownian motion and synthesized multifractal time series. For the former simulation, we find that all methods provide comparable results. For the latter simulation, we perform simulations with an intermittent parameter μ=0.15. We find that the SF underestimates scaling exponent when q>3. The Hilbert method provides a slight underestimation when q>5. However, both DFA and WL overestimate the scaling exponents when q>5. It seems that Hilbert and DFA methods provide better singularity spectra than SF and WL. We finally apply all methods to a passive scalar (temperature) data obtained from a jet experiment with a Taylors microscale Reynolds number Re(λ)≃250. Due to the presence of strong ramp-cliff structures, the SF fails to detect the power law behavior. For the traditional method, the ramp-cliff structure causes a serious artificial energy flux from the low-frequency (large scale) to the high-frequency (small scale) part. Thus DFA and WL underestimate the scaling exponents. However, the Hilbert method provides scaling exponents ξ(θ)(q) quite close to the one for longitudinal velocity, indicating a less intermittent passive scalar field than what was believed before.

Second-order structure function in fully developed turbulence.

We relate the second-order structure function of a time series with the power spectrum of the original variable, taking an assumption of statistical stationarity. With this approach, we find that the structure function is strongly influenced by the large scales. The large-scale contribution and the contribution range are, respectively, 79% and 1.4 decades for a Kolmogorov -5/3 power spectrum. We show numerically that a single scale influence range, over smaller scales is about 2 decades. We argue that the structure function is not a good method to extract the scaling exponents when the data possess large energetic scales. An alternative methodology, the arbitrary order Hilbert spectral analysis which may constrain this influence within 0.3 decade, is proposed to characterize the scaling property directly in an amplitude-frequency space. An analysis of passive scalar (temperature) turbulence time series is presented to show the influence of large-scale structures in real turbulence and the efficiency of the Hilbert-based methodology. The corresponding scaling exponents ζ(θ)(q) provided by the Hilbert-based approach indicate that the passive scalar turbulence field may be less intermittent than what was previously believed.

Scaling of maximum probability density functions of velocity and temperature increments in turbulent systems

In this paper, we introduce a new way to estimate the scaling parameter of a self-similar process by considering the maximum probability density function (pdf) of its increments. We prove this for H-self-similar processes in general and experimentally investigate it for turbulent velocity and temperature increments. We consider turbulent velocity database from an experimental homogeneous and nearly isotropic turbulent channel flow, and temperature data set obtained near the sidewall of a Rayleigh-Benard convection cell, where the turbulent flow is driven by buoyancy. For the former database, it is found that the maximum value of increment pdf pmax(τ) is in a good agreement with lognormal distribution. We also obtain a scaling exponent α≃0.37, which is consistent with the scaling exponent for the first-order structure function reported in other studies. For the latter one, we obtain a scaling exponent αθ≃0.33. This index value is consistent with the Kolmogorov-Obukhov-Corrsin scaling for passive scalar tur...

Analysis of Nonlinear Biophysical Time Series in Aquatic Environments: Scaling Properties and Empirical Mode Decomposition

Aquatic environmental time series often display large fluctuations at many time scales, possessing stochastic properties, as well as deterministic forcing coming from seasonal or annual meteorological and climatic cycles. In this work we are interested in the characterization of these properties, using different statistical tools, borrowed from the field of turbulence, or of nonlinear time series analysis. We first present the analysis of a long (30 years) time series of daily river flow data, recorded in the Seine River (France). We consider the scale dependence and scale invariance of river flow data, using structure function analysis; we also apply a decomposition method called Empirical Mode Decomposition (EMD). We then consider the statistical properties, and the nonlinear dynamics behaviour of a long-term copepod (small crustaceans) time series sampled every week in the Meditarranean sea from 1967 to 1992. We first consider its high variability and characterize its properties, including extreme evens obeying power law tail pdf. We then consider their scale dependence, using Fourier power spectra together with an EMD approach.

Application of arbitrary-order Hilbert spectral analysis to passive scalar turbulence,

In previous work [Huang et al., PRE 82, 26319, 2010], we found that the passive scalar turbulence field maybe less intermittent than what we believed before. Here we apply the same method, namely arbitrary-order Hilbert spectral analysis, to a passive scalar (temperature) time series with a Taylors microscale Reynolds number Reλ 3000. We find that with increasing Reynolds number, the discrepancy of scaling exponents between Hilbert ξθ(q) and Kolmogorov-Obukhov-Corrsin (KOC) theory is increasing, and consequently the discrepancy between Hilbert and structure function could disappear at infinite Reynolds number.

Intermittency measurement in two dimensional bacterial turbulence

In this paper, an experimental velocity database of a bacterial collective motion, e.g., Bacillus subtilis, in turbulent phase with volume filling fraction 84% provided by Professor Goldstein at Cambridge University (UK), was analyzed to emphasize the scaling behavior of this active turbulence system. This was accomplished by performing a Hilbert-based methodology analysis to retrieve the scaling property without the β-limitation. A dual-power-law behavior separated by the viscosity scale ℓ_{ν} was observed for the qth-order Hilbert moment L_{q}(k). This dual-power-law belongs to an inverse-cascade since the scaling range is above the injection scale R, e.g., the bacterial body length. The measured scaling exponents ζ(q) of both the small-scale (k>k_{ν}) and large-scale (k<k_{ν}) motions are convex, showing the multifractality. A log-normal formula was put forward to characterize the multifractal intensity. The measured intermittency parameters are μ_{S}=0.26 and μ_{L}=0.17, respectively, for the small- and large-scale motions. It implies that the former cascade is more intermittent than the latter one, which is also confirmed by the corresponding singularity spectrum f(α) versus α. Comparison with the conventional two-dimensional Ekman-Navier-Stokes equation, a continuum model indicates that the origin of the multifractality could be a result of some additional nonlinear interaction terms, which deservers a more careful investigation.

Analytical Solutions for Composition-Dependent Coagulation

Exact solutions of the bicomponent Smoluchowski’s equation with a composition-dependent additive kernel are derived by using the Laplace transform for any initial particle size distribution. The exact solution for an exponential initial distribution is then used to analyse the effects of parameter on mixing degree of such bicomponent mixtures and the conditional distribution of the first component for particles with given mass. The main finding is that the conditional distribution of large particles at larger time is a Gaussian function which is independent of the parameter .

Numerical study of high-order Lagrangian structure functions in a turbulent channel flow with low Reynolds number

The scaling exponents of Lagrangian velocity structure functions from orders 1 to 10 in a low Reynolds number turbulent channel flow are investigated by using direct numerical simulation. The Reynolds number Reτ is 80 (based on friction velocity on the wall). The Lagrangian velocity structure functions are shown to obey the scaling relations ∼τζL(q). The scaling exponents are normalized by ζL(2) (so-called ESS procedure). The coincidence between the theoretical predictions and numerical calculations is very good for the longitudinal scaling exponent in the channel center. It is also found that the high-order longitudinal scaling exponents agree with theoretical values better than those for the transverse direction.

Large eddy simulation of turbulent statistical and transport properties in stably stratified flows

Three dimensional large eddy simulation (LES) is performed in the investigation of stably stratified turbulence with a sharp thermal interface. Main results are focused on the turbulent characteristic scale, statistical properties, transport properties, and temporal and spatial evolution of the scalar field. Results show that the buoyancy scale increases first, and then goes to a certain constant value. The stronger the mean shear, the larger the buoyancy scale. The overturning scale increases with the flow, and the mean shear improves the overturning scale. The flatness factor of temperature departs from the Gaussian distribution in a fairly large region, and its statistical properties are clearly different from those of the velocity fluctuations in strong stratified cases. Turbulent mixing starts from small scale motions, and then extends to large scale motions.