Page 43 - Computational Fluid Dynamics for Engineers

28 1. Introduction

Y/C
Y/C
0.051-

Airfoil Airfoil
Experimental Ice Shape Q Experimental Ice Shape
NASA, LEWICE 2.0 (Wright) NASA, LEWICE 2.0 (Wright)

_U UJ
(a) x/c (b) - 0.05
X/C
Fig. 1.28. (a) LEWICE prediction of rime ice and (b) LEWICE prediction of glaze ice.

physical model of the ice accretion process. In order to improve our ability to
predict ice shapes corresponding to glaze ice (Fig. 1.28b), it is necessary to (a)
improve the physical model, and (b) compute the heat balance analysis more
accurately. Further details are provided in [5].

1.4.2 Prediction of Aerodynamic Performance Characteristics

The aerodynamic performance characteristics of two- and three-dimensional
bodies can be predicted by either using a Navier-Stokes method (see [22] for
example) or an interactive boundary-layer method (see Chapter 7, [5]) in which
the solutions of inviscid and viscous flow equations are obtained interactively.
Both approaches have merit when applied to airfoils, wings, wing-fuselage and
high-lift systems. Whereas the Navier-Stokes approach offers generality, it is
very computer intensive, requiring considerable run times. Since a viable design
method probably will require the evaluation of many flow conditions, cost is
a major consideration. Furthermore, if attention is focused on predicting the
performance degradation of an aircraft under icing conditions, and consideration
is given to the approximations made in formulating the calculation strategy
for ice shapes and turbulence modeling of flows with iced shapes, the proper
approach becomes clear.
Here the panel method developed by Hess [4, 7], is used to compute the flow-
field about the aircraft. Being a panel method and based on the solution of the
conservation equations without viscous effects, this method does not produce an
accurate prediction of the flowfield about the aircraft, but its accuracy can be
improved by incorporating viscous effects with the procedures discussed in [5].
This method is a very useful engineering tool and the workhorse of computa-
tional methods in industry for aircraft design (Section 1.2). Performing similar
calculations accurately with a Navier-Stokes method is not yet economical.
Since an inviscid method cannot predict the viscous drag, and since the
viscous effects can reduce the lift of the aircraft and its components, it is common