AE Brown Bag
Friday, November 14
11:00 a.m. - 12:20 p.m.
Guggenheim 442
Jai Doshi
Margaret Hwang
Zeina Rmaile
Hayden Bye
Benjamin Jung
Jai Doshi
Title:
ML Modeling of Reactive Flowfields in Solid-Fueled Ramjets
Abstract:
High-fidelity CFD simulations of reactive flowfields are computationally expensive, limiting their use for rapid studies. This work presents a machine learning surrogate for predicting velocity flowfields in solid-fueled ramjet (SFRJ) simulations. A convolutional U-Net was trained on LESLIE CFD data containing grid and transient flowfield data to reproduce flow structures at a reduced spatial resolution. The model predictions accurately capture large-scale structures while significantly reducing computational cost compared to full CFD. Ongoing work aims to extend the surrogate to include physics informed loss, temperature and reaction-rate modeling, super-resolution, and compare performance against other surrogate models. Related studies on hydrogen peroxide detonation and fuel–oxidizer impingement modeling complement this effort by exploring reactive and multiphase flow behavior in high-speed propulsion systems.
Faculty Advisor:
Dr. Suresh Menon
Margaret Hwang
Title:
The Design, Test, and Analysis of Lap Joints in the Characterization of Bolted Joint Loosening
Abstract:
Bolted joint loosening is a critical issue spanning multiple industries and can often lead to complete structural failure. In the past, bolted joint loosening has caused fatalities in incidents such as a train derailment. Loosening occurs when the bolted joint and its system is exposed to a dynamic loading environment, shock, or sustained vibration. Previous research suggests that loosening may not be directly correlated with the specific excitation frequency, and loosening may occur due to a combination of preload and vibration amplitude. As a result, more research is required to characterize the cause of bolted joint loosening. This research focuses on the design, test and analysis of a test fixture to produce repeatable results consistent across multiple tests. The system used consists of two linear oscillators coupled with lap joints. The experiment focuses on varying the contact area and material of the lap joint to characterize how natural frequency and damping ratio changes based on lap joint design. This research presents results found using frequency response functions to characterize natural frequency and compares experimental results to stiffness and damping models presented in previous research.
Faculty Advisor:
Dr. Keegan Moore
Zeina Rmaile
Title:
A Scalable, Modular UAV Platform for Autonomous Flight Research
Abstract:
Advancing autonomous flight research is often constrained by the high cost and limited adaptability of proprietary UAV platforms. To address this, we developed a modular UAV system with a carbon-fiber and 3D-printed structure that enables rapid iteration, sensor reconfiguration, and hardware customization. This lightweight platform provides powerful onboard computing for advanced computer vision and AI algorithms, including simultaneous localization and mapping and object detection. It currently achieves autonomous flight in GPS-denied environments using an off-the-shelf visual–inertial odometry algorithm and will serve as a foundation for future research in multi-robot collaboration.
Faculty Advisor:
Dr. Lu Gan
Hayden Bye
Title:
The Design, Test, and Analysis of Lap Joints in the Characterization of Bolted Joint Loosening
Abstract:
Bolted joint loosening is a critical issue spanning multiple industries and can often lead to complete structural failure. In the past, bolted joint loosening has caused fatalities in incidents such as a train derailment. Loosening occurs when the bolted joint and its system is exposed to a dynamic loading environment, shock, or sustained vibration. Previous research suggests that loosening may not be directly correlated with the specific excitation frequency, and loosening may occur due to a combination of preload and vibration amplitude. As a result, more research is required to characterize the cause of bolted joint loosening. This research focuses on the design, test and analysis of a test fixture to produce repeatable results consistent across multiple tests. The system used consists of two linear oscillators coupled with lap joints. The experiment focuses on varying the contact area and material of the lap joint to characterize how natural frequency and damping ratio changes based on lap joint design. This research presents results found using frequency response functions to characterize natural frequency and compares experimental results to stiffness and damping models presented in previous research.
Faculty Advisor:
Research Engineer Selcuk Cimtalay
Benjamin Jung
Title:
Examining CubeSat Reentry Trajectories through the use of Model-Based Systems Engineering
Abstract:
As CubeSat technology develops, small satellites increasingly transition from pure orbital platforms to those capable of atmospheric reentry for data recovery or technology demonstration. As such, a need arises for complete, integrated, and traceable system modeling of reentry trajectories and possible mission configurations. This research investigated the usage of SysML-based architectures within tools of MagicDraw and Cameo Systems Modeler to develop a digital mission architecture linking mission objectives, complete system requirements, and physical behaviors across a full CubeSat life Cycle. A trajectory suite is developed through MATLAB, beginning with a base modeling of the orbital mechanics and atmospheric reentry dynamics, then with increasingly higher fidelity with the integration of aerodynamic, thermal and structural sub models to build a complete reentry profile. By incorporating a high-fidelity trajectory suite built from scratch with detailed system requirements and components, the study aimed to develop and demonstrate the efficiency of the MBSE methodology. This approach showcased the practicality of MBSE in enabling full mission validation through a comprehensive digital twin, and the benefits of enabling cross-disciplinary engineering integration in a spacecraft mission.
Faculty Advisor:
Research Engineer Selcuk Cimtalay