MULTI-FIDELITY ANALYSIS OF ACOUSTIC STREAMING IN FORCED CONVECTION HEAT TRANSFER

摘要

The behavioral characteristics of thermal boundary layer dictate the relative efficiency of forced convection heat transfer. This research effort is related to the detailed analysis of the temporal evolution of thermal boundary layer under periodic excitations. In presence of oscillations, a distinct thin Stokes layer is formed inside the attached boundary layer, which interacts nonlinearly with the mean flow in the near wall region. This interaction leads to modification of temporally averaged flow fields, commonly known as acoustic streaming. As a result, the aero-thermal wall gradients are modified leading to significant changes in wall shear stress and heat flux. However, the small spatial scales and the inherent unsteady nature of streaming has presented challenges for prior numerical investigations, preventing the identification of optimal parameters. In order to address this void in numerical framework, the development of a three-tier numerical approach is presented. As a first layer of fidelity, a laminar model is developed for fluctuations and streaming flow calculations in laminar flows subjected to travelling wave disturbances. This technique is an extension of the Lin's method to traveling wave disturbances of various speeds (absent of previously employed assumptions), along with inclusion of energy equation. With low computational cost, this level of abstraction is intended to identify the broad parameter space that yield desirable heat transfer alterations. At the next level of fidelity, 2D U-RANS simulations are conducted across both laminar and turbulent flow regimes. This is geared towards extending the parameter space obtained from laminar model to turbulent flow conditions. As the third level of fidelity, temporally and spatially resolved DNS simulations are conducted to simulate the application relevant compressible flow environment. The exemplary findings indicate that in certain parameter space, both enhancement and reduction in heat transfer can be obtained through acoustic streaming. Moreover, the extent of heat transfer modulations is greater than alterations in wall shear, thereby surpassing Reynolds analogy.