A Method for Probabilistic Sensitivity Analysis of Commercial Aircraft Engines
Mavris, Dimitri N.
Roth, Bryce Alexander
Macsotai, Noel I.
MetadataShow full item record
The objective of this paper is to illustrate how probabilistic methods can be utilized to rationally and analytically make design decisions in the presence of uncertainty, with emphasis on the use of probabilistic sensitivities in the aircraft gas turbine engine preliminary design process. A brief review of risk and uncertainty in the engine design process is given, and the role of probabilistic methods is discussed. Probabilistic sensitivity analysis, used in conjunction with response surface methods, is proposed as a computationally-efficient method to address defined sources of uncertainty and risk in engine design from a system level perspective. The method outlined is then applied to the analysis of engine component performance uncertainty impact on the performance of a notional four-engine wide-body commercial transport. More specifically, uncertainty in engine design parameters is shown to have a direct and quantifiable impact on aircraft system figures of merit such as design range and fuel burn. The methods developed are then used to create a set of contour plots showing the behavior of vehicle performance uncertainty over the design space of interest.