Numerical simulation and modelling of non-equilibrium flows in hydro-power applications

点击翻译按钮获取中文摘要

This proposal addresses a problem of critical importance for the hydroelectric power industry, the development of accurate models for the prediction of the flow over rough surfaces, in turbulence far from equilibrium. Two cases will be considered: a separating flow in a curved duct, and the flow in a rough channel undergoing system rotation. We will use direct and large-eddy simulations (DNS and LES) of turbulence to validate the models for the Reynolds-Averaged Navier-Stokes equations that are commonly used in industry. Although the DNS and LES will be performed at Reynolds numbers lower than those in actual applications, the LES should be able to achieve the fully rough regime that allows to extend the simulation results to practical applications. The proposed research will benefit Canada in several ways: first, it will train one PhD student and one PDF in the computational sciences; secondly, it will generate fundamental knowledge on the physics of rough-wall boundary layers in the presence of swirl, curvature, system rotation and separation. Finally, it will develop prediction models that will benefit the power generation industry (and may be extended to other areas of application), and can potentially result in significant efficiency increases and financial savings.