Estudo da transição laminar-turbulenta utilizando CFD

Abstract

In fluid mechanics the boundary layer is the region of the flow that senses a direct influence by the presence of the limiting surface. This concept can be also applied for an atmospheric flow. The flow on the atmospheric boundary layer is predominantly turbulent, however, after the sunset, the radiation incidence caused by the sun cease and the surface starts to be cooled down by a longwave radiation emission. Therefore, an adjacent air layer is also cooled down, resulting in a thermally stratified boundary layer. On this stratified boundary layer the turbulence is predominantly mechanic, in night with strong stratification, usually night with a clean sky, weak wind and big radioactive loss, the turbulence activity decreases by several orders of magnitude, but may reappearing abruptly and unpredictably in time and space. A numeric model of a flow in these conditions is a very hard task, the most model in the existing literature cannot describe the unpredictable behavior of the flow. Resent research suggests that this problem can be caused by the way that the diffusivity of the eat and momentum are described on the stratified boundary layer and are related to each other. Usually, the momentum diffusivity coefficient is estimated by a prognostic equation and the eat diffusivity can be obtained with the results of this equation, taking into consideration the turbulent Prandtl number of the flow. Many analyses show that the Prandtl number is virtually constant and close to the unity, in weak atmospheric stability, what is utilized by the most existing atmospheric models. However, recent researches suggest two distinct regimes that depend of the Prandtl number with the atmospheric stability, indicating strong dependence with each other. This way, this work aims the development of a numeric experiment to the study of the laminar-turbulent transition, using a computational fluid dynamic shareware where the surface stratification will be ignored, because in this stage of the work the only interest is the surface dimensions of the domain utilized, the minimum velocity to the flow transition to occur and how this transition affects the momentum and eat energy diffusivity on the flow.