At first glance, preliminary aircraft engine design
may seem well-defined and, therefore, in little need of
probabilistic methods. After all, accurate predictions for the
performance of engine components as well as of the overall system
are possible using existing analysis techniques. However, the
cumulative effect of the many uncertainties in engine component
performance does impact overall system performance.
For a theoretical large commercial aircraft, it is not unusual
to have a cumulative design uncertainty of 5%. While the
likelihood of worst-case extremes is small, variation of 100 n mi
on either side of the mean design flight range is possible. In
today's highly competitive marketplace, this is significant enough
to warrant further consideration.
Typically, design margins based on hard-won experience
compensate for uncertainties in engine performance estimates.
However, much interest within the aircraft engine industry is
centered on robust and probabilistic methods. This interest is
driven by increased competitive pressures, demand for greater
safety, and longer mean time between failures, environmental
consciousness, and maturation of the engine and associated
technology.
The first three points make engine design more difficult, with
design freedom increasingly limited as time goes on. Technology
maturation has slowed the pace of major developments over the past
decade. As progress slows and constraints become more restrictive,
engine designers must extract every bit of performance from
current technologies while simultaneously satisfying all
requirements.
If one accepts this technology maturation argument, designers
of future engines will have to find ways of getting superior
performance without the benefit of major technological advances.
The way to accomplish this is by refining current designs and
trimming design margins while staying within safety requirements.
Most critical decisions are made in the early stages of
development where available design freedom can best achieve better
performance. Probabilistic design provides an analytical framework
that allows the designer to improve performance by determining the
necessary design margin, the parameters impacting the uncertainty
in performance, and ways to reduce the impact of uncertainty.
Probabilistic methods have been applied by researchers at the
Georgia Institute of Technology to the preliminary design process
for a high-bypass engine as installed on a theoretical,
400-passenger, commercial aircraft. Engine and mission performance
figures of merit (FoMs) are tracked to show the impact of changing
the cycle parameters of a scaleable, fixed-configuration engine on
a fixed-size, four-engine aircraft's performance.
Information was provided by Dimitri N. Mavris, Noel I.
Macsotai, and Bryce Roth of the Georgia Institute of Technology.
