| The application of optimization to spacecraft design has the potential to significantly improve decision-making capabilities early in the design process. Successful integration of optimization with system-level spacecraft tools allows the concept architecture, technology choices, and performance requirements to be adjusted to meet an overall estimated cost goal, thereby enabling spacecraft design to be completed in a reliable, economical and timely manner.; As increasingly sophisticated synthesis tools provide more realistic models of the spacecraft system, it is useful to investigate the ability of different optimization techniques to operate on these mathematical models. This research provides an evaluation of optimization techniques as they are applied to increasingly complex spacecraft design problems. Specifically, this research investigates the advantages and disadvantages of traditional, closed-form optimization algorithms and non-traditional heuristic methods on system-level spacecraft models. Performance comparison includes the ability of these methods to efficiently find optimal and near optimal solutions in situations where exhaustive enumeration is computationally impractical.; The principal contribution of this thesis is the introduction of an optimization methodology integrating disciplinary and system-level spacecraft models, optimization techniques, and sensitivity analysis into a single computational flow. The methodology provides a general-purpose analytical framework for efficiently assessing the impact of disciplinary technology trade options on system-level performance and cost goals, thereby significantly increasing decision-making capabilities early in the design process. |
