Application of Trade Space Visualization to Discrete and Continuous Complex Dynamical Systems

Open Access
- Author:
- Jordan, Daniel
- Graduate Program:
- Aerospace Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- None
- Committee Members:
- Dr David B Spencer, Thesis Advisor/Co-Advisor
Dr David B Spencer, Thesis Advisor/Co-Advisor - Keywords:
- differential evolution
visual steering
dynamical systems
constant acceleration
constant thrust
continuous thrust
trade space
human-in-the-loop optimization - Abstract:
- Engineering design problems often contain correlations and trade-offs that may or may not be obvious or well-understood. As design problem complexity increases, decision makers find it more and more difficult to grasp these trade-offs effectively. The rapid growth of computing power now allows the simulation of millions of design alternatives. Understanding the trade-offs associated with these alternatives has never been more important. Trade space visualization tools are being developed to aid decision makers by allowing them to effectively explore a design space and identify the underlying trade-offs and nuances particular to a specific problem. These tools provide great potential in evaluating complex dynamical systems in the aerospace industry, among others. This work explores the application of the Applied Research Lab Trade Space Visualizer (ATSV) to various discrete and continuous complex dynamical systems. First, the motivation for and capabilities of ATSV are covered. Then, ATSV is applied to two test problems: (1) a nonlinear mass-spring-damper system and (2) a simple two-burn impulsive spacecraft maneuver. Optimal Pareto solution sets are obtained for each system. Finally, the application of ATSV to more complicated continuous thrust spacecraft maneuvers is considered. The identification of a known optimal solution is followed by modification of the equations of motion to include a discrete hardware-side design variable, namely, engine type. Optimal solutions are located for each corresponding design variable option. A study is performed to understand the sensitivity of optimal trajectories to perturbations of the initial orbit. For each problem explored in this work, qualitative observations are made on the effectiveness of ATSV as a solution acquisition tool with particular emphasis on ATSV’s visual steering capabilities.