A Methodology for Identifying Experiments For Uncertainty Mitigation in Complex Multi-Disciplinary Design
Author(s)
Yarbasi, Efe Yamac
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
The design of a flight vehicle is a lengthy, expensive process spanning many years. Thanks to increasing computational capabilities, designers have been relying on computer models to make predictions about the real-life performance of an aircraft. However, the results obtained from computational tools are never exact due to a lack of understanding of physical phenomena, inadequate modeling, and abstractions in product details. The vagueness in quantities of interest is uncertainty. Because most of the cost is committed early in the design, any decision made on quantities involving significant uncertainty may result in budget overruns, schedule delays and performance shortcomings, as well as safety concerns. The goal of this thesis is to develop a systematic methodology to identify and mitigate the sources uncertainty in aircraft design, with a focus on uncertainties due to a lack of knowledge epistemic, namely model-form and parameter uncertainties.
An aircraft is a complex, multi-disciplinary system that is built as integration of other intricate subsystems. To make sure that all subsystems and the integrated system meet the pre-defined requirements, Systems Engineering (SE) practices are widely adopted throughout the aerospace industry. However, SE methods fall short of accounting for the implications of using a specific modeling and simulation environment, and simulation-borne uncertainties. The first objective of this thesis is to enhance SE by providing a way to incorporate components of a modeling and simulation activity while adhering to established principles. The second research area of this thesis addresses some of the prominent issues faced in identifying critical uncertainties, so that resources can be allocated to uncertainties that would make the biggest impact on the design. After the critical uncertainties are identified, computational and/or physical experiments can be designed to create new information, so that any epistemic uncertainty can be reduced. However, real-life operational conditions cannot be exactly duplicated in tests due to many reasons such as testing facility constraints. The third and final research area addresses the identification of optimal experiment conditions for computational and physical experiments such that real-life operational conditions can be best approximated.
For each focus area, the solution approaches to each research question will be demonstrated on an appropriate, self-contained problem. The cumulative output of this thesis will be a complete, four-step methodology that can be tailored for a specific application, effectively guiding it, and highlighting the pitfalls to avoid.
Sponsor
Date
2023-08-08
Extent
Resource Type
Text
Resource Subtype
Dissertation