Developments in Linear Aeroelastic Gust Sensitivities for Gradient-Based Shape and Sizing Optimization

Date of Award

8-15-2026

Degree Name

Ph.D. in Aerospace Engineering

Department

Department of Mechanical and Aerospace Engineering

Advisor/Chair

Markus Rumpfkeil

Abstract

Aircraft design is often broken down into three phases: conceptual, preliminary, and detailed design. With each phase, additional data is produced to refine the performance and design details of the aircraft. Often, many key design decisions are made in the conceptual design phase where limited information is available but restricts the design freedom in later phases of design. As the design of an aircraft progresses, the cost to correct `mistakes' grows, especially as the attributes of an aircraft are realized. Many examples of deficient designs exist, from subsystem integration, thermal loads, and aeroelastic instabilities. This body of work seeks to address aspects of gusts in aeroelastic design and enable its inclusion into earlier phases of design. In this dissertation, a couple of methods are explored to efficiently capture the effects of an aeroelastic response due to a gust using conceptual and preliminary design analysis methods. Starting with the Equivalent Static Load method, linear aeroelastic gust responses from NASTRAN are used to determine critical loads and then optimized using NASTRAN Solution 200. This enables the inclusion of gust loads in a commercial aeroelastic design tool, as well as the `easy' addition of other aeroelastic analysis methods such as flutter and quasi-static maneuver loads. A key piece that is missing from NASTRAN is the ability to vary the shape of the vehicle. To solve this shortcoming, the NASTRAN-based optimizations are implemented as sub-optimizations in a Sequential Linear Programming algorithm, allowing for variations in the shape of the vehicle using finite differences. While this method is straightforward to implement and a robust approach to include gust loads in aeroelastic design, it is computationally expensive to assess the sub-optimizations. It also does not address analytic shape sensitivities, which are critically important in early stages of design. The remainder of the dissertation lays out a method to include analytic shape and sizing sensitivities for aeroelastic gust optimization. The mode displacement method is used to develop the aeroelastic equations of motion for a gust response and recover the stresses for a von Mises stress constraint. The von Mises stress constraint is then differentiated to provide sensitivities suitable for optimization. A key component for accurate analytic sensitivities is the inclusion of eigenvector sensitivities. A simple tapered wing and an X-56A-like wing are optimized to understand the impact of a gust on the design and how different gusts frequencies change the wing design. Additionally, the role of inertial relief is explored in the response of gusts. The results of the optimization studies for both models, the tapered and the X-56A-like wing highlight the importance of gust loads on design. In the case of the tapered wing, neither the static aeroelastic maneuver load nor the gust load was sufficient to design the wing alone. In order for a feasible design, both maneuver and gust loads must be included in the design. Likewise, in order for the X-56A-like wing to withstand `off-design' gust loads, additional gust constraints are required. The more gust loads that are included in the design problem, the less detrimental the `off-design' gust loads become.

Keywords

Aerospace Engineering

Rights Statement

Copyright 2026, author

Link to Full Text

Share

COinS