Design coupling analysis in multidisciplinary design optimization and control co-design

The design of complex engineering systems involves multiple interconnected, or coupled, disciplines. The relationships between these disciplines are not always well-known, especially for emerging technologies. Optimizing these systems can be a challenging task, especially if the designer has limited knowledge of the system at hand. One of the questions that arises early in the optimization problem formulation is what design variables to optimize. Complex systems may have hundreds of design variables, and it might not be possible or even desirable to optimize them all. More clear understanding of relationships between problem variables can aid optimization specialists in value-added problem definition. This work focuses on extracting valuable insights from design coupling analysis to make optimization problem formulation decisions, focusing on design variable selection. These decisions may be based on expert intuition, but important problem formulations may be missed by relying solely on intuition. This work focuses on multidisciplinary design optimization (MDO) and control co-design (CCD) problem formulation. By performing design coupling analysis (DCA), problem formulations that reduce the computational expense associated with solving MDO and CCD problems can be identified. Furthermore, the use of design coupling analysis may identify non-obvious problem formulations and provide design insights that lead to new engineering system designs. Both MDO and CCD present a number of challenges for the study of design coupling, making it necessary to develop new methods to estimate coupling. This dissertation expands existing work for design coupling estimation in MDO by considering together the effect of design variables on the objective function, and on each other. Previous work in design coupling estimation in CCD imposed some restrictive assumptions, such as considering a limited view of coupling or assuming that the control variables are scalar quantities. Here, many of those restrictions are lifted. The design coupling methods proposed in this work allow for bidirectional coupling, that is, plant and control affect each other, and for time-varying control inputs. The proposed work is investigated in three case studies spanning two rich applications, spacecraft design and attitude control, and floating offshore wind turbines. The case studies are used to demonstrate the quantification of design coupling and the use of this information to formulate optimization problems.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Aerospace Engineering

Grantor

University of Illinois Urbana-Champaign

Year dc:date

2025

Author dc:creator

• Fernandez Bravo, Elena

Contributors dc:contributor

• Allison, James T
• Ornik, Melkior
• Wang, Pingfeng
• Coverstone, Victoria

• Design coupling
• Multidisciplinary design optimization
• Control co-design
• Optimization
• Optimization problem formulation
• Spacecraft control
• Floating offshore wind turbine

Statement dc:rights

• Copyright 2025 Elena Fernandez Bravo

Language dc:language

en

Handle dc:identifier

https://hdl.handle.net/2142/132533

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/132533