ü From 8 September – 13 October 2025 (5.5 Weeks, 11 Classes, 33 Total Hours)

ü Every Monday and Wednesday at 12–3 p.m. Eastern Time (UTC-5) (all sessions will be recorded and available for replay; course notes will be available for download)

ü This new essential course covers reduced-order model (ROM) techniques, focusing on aircraft performance and aerodynamic load analysis

ü All students will receive an AIAA Certificate of Completion at the end of the course

OVERVIEW

This course is designed to provide a comprehensive introduction to the application of reduced-order model (ROM) techniques in aerospace engineering, focusing on aircraft performance and aerodynamic load analysis. The ROM techniques are classified into two categories: data-driven methods based on machine learning (ML); and equations-derived methods based on nonlinear projections (NP). The course aims to equip students with a foundational understanding of ML and NP concepts and practical skills for implementing these approaches in aerospace contexts. By the end of the course, participants will gain knowledge of cutting-edge techniques and their relevance to complex aerospace problems, particularly in predicting steady and unsteady aerodynamic loads, and performing aeroelastic analysis.

The course begins with an introduction to core aerospace engineering problems, focusing on aircraft performance and the complexities involved in aerodynamic and aeroelastic load prediction. There is an overview of steady-state and unsteady aerodynamic modeling approaches. These sessions cover critical topics such as the aerodynamic characteristics of steady and transient flows around aircraft, which are essential for understanding the needs of ROM approaches to accelerate the rate of the design process.

A significant portion of the course is devoted to ROM techniques, which simplify complex aerodynamic and aeroelastic problems into manageable computational tasks. Students will learn about two primary types of ROMs: data-driven ROMs that leverage machine learning algorithms and equations-derived ROMs that rely on projection-based methods. The data-driven ROM segment introduces machine learning techniques such as autoencoders and graph neural networks, highlighting their advantages in handling large datasets with minimal computational cost. Meanwhile, the projection-based ROM session covers techniques such as NP of the residual of the coupled system, which reduces computational complexity without compromising accuracy.

Throughout the course, real-world aerospace applications are emphasized to ensure that participants develop skills that are directly transferable to industry settings. By blending ML techniques with aerospace engineering fundamentals, this course provides a unique perspective on modern aerospace challenges and prepares participants to apply innovative solutions in their careers.