Mission-specific design of aircraft energy systems

Aksland, Christopher Thomas

Permalink

https://hdl.handle.net/2142/120262 Copy

Description

Title

Mission-specific design of aircraft energy systems

Author(s)

Aksland, Christopher Thomas

Issue Date

2023-04-18

Director of Research (if dissertation) or Advisor (if thesis)

Alleyne, Andrew

Doctoral Committee Chair(s)

Mehta, Prashant

Committee Member(s)

• Beck, Carolyn
• James, Kai
• Clark Jr., Daniel

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Energy Systems
• Multi-Disciplinary Design Optimization
• Topology Optimization
• Optimal Control
• Experimental Validation

Abstract

The promise of electrified aircraft aims to revolutionize the transportation industry as a more sustainable and capable mode of modern travel. However, the current power and energy density of these systems is lacking, resulting in aircraft with significantly limited range in comparison to their traditional fuel-powered counterparts. This limitation is partially an artifact of the complex aircraft energy systems that are comprised of multi-domain and multi-timescale dynamics with strict operational limitations. Additionally, accounting for such constraints and complex dynamic interactions can be challenging when also considering the wide range of potential applications for electrified aircraft. To address these challenges and improve power and energy density, one approach is more intelligent system operation; make optimized control decisions that more effectively manage and utilize onboard energy resources. This dissertation proposes a multi-timescale hierarchical model predictive control approach for the operation of the multi-domain aircraft energy systems that inherently accounts for system coupling and operational constraints in the control decision-making process. A key element of the control design is the creation and integration of an optimization-oriented graph-based modeling framework, designed specifically for conservation systems. This graph modeling framework facilitates computationally efficient control design by decomposing the optimal control problem into a hierarchy of coordinated sub-problems. The hierarchical control approach is experimentally validated on an experimental testbed that represents an aircraft integrated power, propulsion, and thermal management system. In comparison to a conventional control approach, the hierarchical controller demonstrated advancements in performance and efficiency metrics with significant improvements in operational reliability. While this demonstration established that control alone can improve a system's performance, it is also necessary to consider the additional flexibility and effectiveness gained from plant design. Combined plant and control design (co-design) offers an approach to optimal system-level aircraft design by considering architecture, sizing, and control design disciplines. Together, these design methodologies comprise the components/technologies integrated into the aircraft design, the connections between those components to build the system, and their operation. This dissertation focuses on integrating feedback control design within architecture and sizing optimization processes. The design methods leverage the graph-based modeling approach to efficiently encode the plant dynamics into the optimization routine. Modular graph models play a key role in the architecture optimization process because they enable the evaluation of a variety of system topologies. A scalable combined plant sizing and feedback control optimization is applied to a hybrid UAV design, demonstrating significant improvements in closed-loop system performance in contrast to other conventional design strategies. Additionally, a novel and efficient, relaxed approach to the architecture optimization problem was used to create mission-specific plant and feedback control designs for a thermal management system. The result identified a system topology that is optimized for a range of operating requirements. Notably, the proposed approach could identify near optimal system designs an order of magnitude faster than traditional exhaustive search approaches.

Graduation Semester

2023-05

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/120262

Copyright and License Information

Copyright 2023 Christopher Aksland

Owning Collections