PROGRESS IN DIRECT NUMERICAL SIMULATION OF TURBULENT HEAT TRANSFER

Nobuhide Kasagi, Oaki Iida

OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 1999 · 52 citations · 77 references

Concepts

Abstract

With high performance computers, reliable numerical methods and efficient post-processing environment, direct numerical simulation (DNS) offers valuable numerical experiments for turbulent heat transfer research. In particular, one can extensively study the turbulence dynamics and transport mechanism by visualizing any physical variable in space and time. It is also possible to establish detailed database of various turbulence statistics of turbulent transport phenomena, while systematically changing important flow and scalar field parameters. The present paper illustrates these novelties of DNS by introducing several examples in r ecent studies. Future directions of DNS for turbulence and heat transfer research are also discussed. In the following, with introductory remarks on computational requirement for respectable DNS, progress in numerical techniques is first reviewed. Then, an effort to reveal the effect of Prandtl number on turbulent heat transfer is sketched with an emphasis on the new techniques developed for making DNS over a wide range of Prandtl number possible. Both methods of saving grid points and Lagrangian particle tracking are found very useful in the future work on turbulent heat and mass transfer. The complex buoyancy effects on turbulent transport are discussed in a series of DNSs of horizontal and vertical channel flows. It is found that the buoyancy effect appears much differently depending upon the directional alignment of the buoyant force and the mean flow. In the vertical flow, the buoyancy effect results in an increased or decreased Reynolds number effect, which appears through a change in the Reynolds stress distribution and hence in the shear production rate of turbulent kinetic energy near the wall. In the horizontal flow, however, it causes substantial alternation in the turbulence structures and transport mechanism so that the resultant heat transfer coefficient would behave in a much more complex manner. The thermal plumes and internal gravity waves characteristic of these flows play an important role in the interaction with the quasi-coherent structures of wall turbulence. DNSs of the channel flow under active turbulence and heat transfer control are discussed finally. In these simulations, an array of micro deformation actuators on a wall surface or intelligent micro particles are tested as control schemes. It is shown that these sophisticated devices possibly bring about notable friction drag reduction and heat transfer enhancement. A perspective view is given that DNS will be even more useful in evaluating future turbulence control methodologies based on new algorithms such as optimum control theory and neural networks. These simulations have been triggered by the rapid development of microelectromechanical systems (MEMS) technology, but hopefully future DNSs will lead further development of MEMS-based controller units integrating micro sensors, micro actuators and IC.

References

77