Generation of Turbulent Inlet Conditions for Thermal/Velocity Boundary Layer Simulations

A new approach for simulating realistic turbulent temperature information at the entrance o f a computational domain which considers a spatially evolving turbulent boundary layer is presented. According to the method proposed by Lund et al . [3 ], turbulent velocity data is predicted from an auxiliary simulation. In the development section prior to the computational domain, the flow field is spatially growing. Here, the above mentioned subsidiary simulation generates inflow conditions through a sequence of simple operations where the velocity field at a downstream station, the so called recycle plan e, is rescaled using empirical scaling and re -injected at the inlet of the development section until reaching some convergence criterion. The same concept was employed by Kong et al . [2] for predicting turbulent inflow temperature fluctuations in Direct Nu merical Simulations (DNS) of a flat plate with isothermal and isoflux wall boundary conditions. Since the previous models use a single scaling as in the classical approach, not only for the velocity but also for the temperature fields; this research propos es new scales determined from the analysis of the RANS (Reynolds Averaged Navier – Stokes) equations for the inner and outer regions. The velocity scales are based on the work of George and Castillo [1], meanwhile the temperature scales are able to absorb the effects from the pressure gradient, upstream conditions and the Peclet number dependence are a modified form of the equilibrium similarity analysis of Wang and Castillo. In this way, a broader range of turbulent inflow thermal fluctuations may be gene rated in the future , particularly; for pressure gradient flows . The new models are te sted in DNS over a flat plate with a zero pressure gradient. The rescaling process is done at each time step in the main flow simulation; in other words, our model does no t consider an auxiliary simulation.