Project Title:

A New Approach to the Prediction of Transition in Gas Turbines

94-1		02.01	2600			AMOUNT REQUESTED $

A New Approach to the Prediction of Transition in Gas Turbines

Abstract:

The objective of this proposal is to develop advanced analysis 
software based on three-dimensional computational fluid dynamic 
(CFD) methods to support the design and optimization of gas 
turbines. 
In fact, due to the process of transition from a laminar to a 
turbulent boundary layer on the blade, great uncertainty presently 
exists in predicting convective heat transfer rates from the blade. 
Current CFD techniques using inadequate transition criteria, based 
on local pressure gradient and boundary-layer thickness parameters, 
fail to address the fact that the competing effects of high 
external stream turbulence and severe streamwise acceleration can 
lead to the boundary layer being in an intermediate state between 
laminar and turbulent over much of the blade surface. We propose to 
use a well-validated procedure, based on state-of-the-art 
turbulence modeling, for predicting momentum and heat transfer 
rates in boundary layers with especial reference to the transition 
from laminar to turbulent (or semi-turbulent) flow induced by high 
levels of external stream turbulence.  
Phase I will focus on demonstrating the innovative technology and 
assessing our turbulence models in the design of gas turbines. This 
will be accomplished by modeling the development of a boundary 
layer on the blade surface under the influence of various free 
stream turbulence energy levels. The state-of-the-art in turbulence 
modeling will include the use of low-Reynolds number RNG-based 
eddy-viscosity models, non-linear k-e, explicit algebraic stress 
models and other alternatives to a full low-Re second-moment 
closure.  
Phase II will focus on improving and enhancing physical models as 
deemed necessary in Phase I, as well as employing the use of 
sophisticated full Reynolds-stress closures that incorporate 
sufficiently general sub-models of the different processes at work 
for there to be good prospects of it predicting transition with 
reasonable accuracy.  

The proposed work employs many basic features that will find wide 
usage in industry. The code developed may be used for modeling 
blade-cascade flows over a wide range of operating conditions and 
will be a valuable tool for the design and optimization of gas 
turbines. 

KEY WORDS

Fluent Inc.
10 Cavendish Ct.
Lebanon, NH 03766