Doctoral Defense: John W. Dykes

Primary tabs

You are invited to a
Ph.D. Thesis Defense

by

John W. Dykes
Advisor: Prof. Dimitri Mavris
9 a.m., Monday, August 8
Weber Building - Collaborative Visualization Environment (CoVE)

TEMPUS: A Methodology For Model-Based Robust-Optimal Design of Time-Dynamic System Identification Experiments Using Variational Asymptotic Expansions

 

ABSTRACT:
The development of appropriate flight tests has proven to be a critical piece in the design process of many revolutionary next-generation aerospace vehicles.  For example, in the case of hypersonic vehicles with air-breathing SCRAMjet engines, sophisticated computational analyses have been developed, requiring extensive validation and calibration. The current state of ground testing facilities has not been able to accommodate these demands, due to the inability to replicate hypersonic flow conditions with sufficient accuracy, and these deficiencies have put more demand and pressure on flight experiments, which have historically proven to produce the highest quality results but at the potential price of extreme complexity and expense.  As observed in the case of hypersonic SCRAMjet vehicle flight testing, high expense has led to conservative, risk-averse experiments which often yield little gain in knowledge.

There is an entire discipline devoted to the process of design and information extraction from aerospace-type experiments known as aircraft system identification (SysID), which fuses three strongly interdependent topics: (1) instrumentation and measurement systems, (2) experiment design, and (3) statistical estimation. At its essence, SysID attempts to develop time-dynamic experiments so that statistical estimation techniques can most effectively be used to identify high-confidence physics-based models. An implicit limitation to this process lies within the topic of optimal experiment design (OED), which is often posed as an optimal control problem for the concurrent design of aircraft maneuver inputs, instrumentation system parameters, flight conditions, test duration, etc. Here, Fisher information-based optimality criterion are sought to be utilized; however, these metrics can only be accurately computed if the true values of the unknown models and parameters are known prior to conducting the actual experiment. Commonly referred to as the circulatory problem in statistics literature, OED often requires an augmented robust-optimization approach (ROED) to account for inaccuracies due to parametric uncertainty.
 
The focus of this research is to consider the design of flight-dynamic experiments from a systems engineering perspective, combining elements of SysID and ROED to concurrently design information-dense robust-optimal flight experiments, in the presence of modeling uncertainties. The proposed methodology is called TEMPUS, which stands for Time-dynamic Experiment design using a Model-based approach to Propagate Uncertainty for System identification, and will be demonstrated using the Generic Hypersonic Vehicle (GHV) model to design information-dense SCRAMjet-powered flight tests. A notable gap that this methodology must overcome, is the automated and accurate computation of variational asymptotic expansions (i.e. higher-order dynamic state-to-parameter sensitivities), which are necessary to compute the information (robust)-optimality metrics. To address this gap, an automatic differentiation tool that is specialized for use in robust-optimal dynamic experiment design is developed.
Committee Members:
Prof. Rafael de la Llave
Dr. Charles Domercant
Dr. Jon Zumberge
Prof. Daniel Schrage

Groups

Status

Categories

  • No categories were selected.

Keywords

  • No keywords were submitted.