Three-Di mensional Computations of Multiple TandemJets in Crossflow

The mixing and merging characteristics of multiple tandem jets in crossflow are investigated by use of the Computational Fluid Dynamics （CFD） code FI,UENT. The realizable k - ε model is employed for turbulent elosure of the Reynolds-averaged Navier-Stokes equations. Numerical experiments are performed for 1-, 2- and 4-jet groups, tbr jet-tocrossflow velocity ratios of R = 4.2 ～ 16.3. The computed velocity and scalar concentration field are in good agreement with experiments using Particle Image Velocimetry （PIV） and Laser Induced Fluorescence （LIF）, as well as previous work. The results show that the leading jet behavior is similar to a single free jet in crossflow, while all the downstream rear jets have less bent-over jet trajectories - suggesting a reduced ambient velocity for the rear jets. The concentration decay of the leading jet is greater than that of the rear jets. When normalized by appropriate crossflow momentum length scales, all jet trajectories follow a universal relation regardless of the sequential order of jet position and the nund）er of jets. Supported by the velocity and trajectory measurements, the averaged maximum effective crossflow velocity ratio is computed to be in the range of 0.39 to 0.47.

Numerical Experiment on Two-Dimensional Line Thermal

The time evolution of a two-dimensional line thermal-a turbulent flow produced by an initial element with significant buoyancy released in a large water body, is numerically studied with the two-equation k - epsilon model for turbulence closure. The numerical results show that the thermal is characterized by a vortex pair flow and a kidney shaped concentration structure with double peak maxima; the computed flow details and scalar mixing characteristics can be described by self-similar relations beyond a dimensionless time around 10. There are two regions in the flow field of a line thermal: a mixing region where the concentration of tracer fluid is high and the flow is turbulent and rotational with a pair of vortex eyes, and an ambient region where the concentration is zero and the flow is potential and well-described by a model of doublet with strength very close to those given by early experimental and analytical studies. The added virtual mass coefficient of the thermal motion is found to be approximately 1. The aspect ratio for the kidney-shaped sectional thermal is found to be around 1.45 for the self-similar phase. The predicted thermal spreading and mixing rate compares well with experimental data.

Stirring by anisotropic squirming

We consider a fluid stirred by the locomotions of squirmers through it and generalize the stochastic hydrodynamic model proposed by Thiffeault and Childress,Phys.Lett.A(2010)and Lin et al.,J.Fluid Mech.(2011)to the case in which the swimmers move in anisotropically random directions.A non-diagonal effective diffusivity tensor is derived with which the diffusive preference of a passive particle along any given direction can be computed to provide more details of the phenomena beyond scalar statistics.We further identify a fraction from the orthogonal decomposition of the drift-induced particle displacement to distinguish the underlying nonlinear mixing mechanism for different types of swimmers.Numerical simulations verify the analytical results with explicit examples of prescribed,anisotropic stirring motions.We also connect our formulation to several measures used in clinical medical research such as diffusion tensor imaging where anisotropic diffusion has a significant consequence.