Harok Bae, Ph.D. (Advisor); Ramana V. Grandhi, Ph.D. (Committee Member); Rory A. Roberts, Ph.D. (Committee Member); Markus P. Rumpfkeil, Ph.D. (Committee Member); Edwin E. Forster, Ph.D. (Committee Member)
Doctor of Philosophy (PhD)
To make coupled multi-physics-informed design decisions, multidisciplinary analysis, design optimization and uncertainty quantification must be present to accurately represent the full system under investigation. Unfortunately, all of these processes are computationally demanding, requiring a large number of system evaluations with identified uncertain variables, and iterative system evaluations with respect to the design variables of interest. Surrogate or metamodels are used to alleviate the computational burden in both these design exploration activities by trading accuracy with efficiency. The primary objective of this dissertation is to develop a flexible surrogate modeling technique capable of quantifying the uncertainty of multidisciplinary systems in an iterative and efficient procedure. In this work, the Non-Deterministic Kriging (NDK) method is derived. This surrogate model represents a flexible approach for approximating epistemic and aleatory uncertainty. To achieve an iterative and efficient computational framework additional tasks were established: (1) characterize and develop a unified stochastic process incorporating incomplete and mixed uncertainty data; (2) develop a novel adaptive sampling method that effectively and efficiently updates the NDK model to enable a global multidisciplinary design optimization technique under uncertainty; (3) derive analytic sensitivities to achieve non-deterministic sensitivities with respect to the design variables; (4) propose an efficient reliability-based design optimization framework for multidisciplinary systems using NDK to reduce the design space.
Department or Program
Ph.D. in Engineering
Year Degree Awarded
Copyright 2019, some rights reserved. My ETD may be copied and distributed only for non-commercial purposes and may not be modified. All use must give me credit as the original author.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.