AE Brown Bag Seminar
Friday, April 5
11:00 a.m. -1:00 p.m.
Guggenheim 442
Pizza Served
Madeleine Barbe
Palin Bhardwaj
Mihnea Pop
Maria Cayetana Salinas Rodriguez
Nathan Wong
Madeleine Barbe
Title:
Wind Tunnel Analysis of Landing Gear Drag to Improve Small Scale Aircraft Performance
Abstract:
Drag analysis composes a significant portion of Georgia Tech’s Design, Build, Fly competition team. The drag of a vehicle has significant contributions to the endurance and top speed of an aircraft, which are vital components to attaining a high score within various flight missions. This project is focused on verifying the empirical methods of landing gear drag analysis used historically within the club using wind tunnel analysis of landing gear and wheel test pieces. Furthermore, recommendations of improved landing design may be made based on drag analysis of the landing gear test pieces.
Faculty Advisor:
Professor Carl Johnson
Palin Bhardwaj
Title:
Applications of SysML Modeling for Aerospace System Design
Abstract:
Model-Based Systems Engineering (MBSE) is an approach to designing complex systems by creating models representing many aspects of the system, including its requirements, structure, behavior, and interactions. Combining Multi-Disciplinary Analysis and Optimization (MDAO) with MBSE is very helpful for aerospace systems design. Systems Modeling Language (SysML) has become a very common modeling language for MBSE that helps create and manage intricate system models. Using SysML, we are also able to verify and validate our models to ensure maximum consistency with results. The integration between MBSE and MDAO help with model consistency and allow for advanced analysis and optimization. For this project, using MagicDraw, we are able to model complex aerospace systems, specifically focusing on SpaceX's Falcon 9 launch vehicle. The focus is representing vital systems of the vehicle while integrating various tools in SysML to have a strong representation of model-based systems engineering.
Faculty Advisor:
Professor Selcuk Cimtalay
Mihnea Pop
Title:
Jet in Cross Flow CFD Automation
Abstract:
The research summary presents an approach to automate the process of Computational Fluid Dynamics (CFD) simulations and post-processing, for the analysis of cross flow phenomena. Completed initially in Ansys Fluent, automated using PyFluent and post-processing done in Tecplot utilizing PyTecplot for automation. The research aims to achieve full automation, streamlining the workflow from mesh setup to result visualization by leveraging PyFluent for ANSYS Fluent automation and PyTecplot for post-processing tasks. Repetitive manual tasks are eliminated, reducing human intervention and potential errors and allowing for multi-processing or iterative adjustments to simulations. The integration of automation tools enhances efficiency, allowing for rapid iteration and analysis of simulations while allowing users to quickly change parameters without reconstructing the workflow manually.
Faculty Advisor:
Professor Timothy Lieuwen
Maria Cayetana Salinas Rodriguez
Title:
Controlling an aircraft’s longitudinal dynamics using a PID controller
Abstract:
A PID (Proportional-Integral-Derivative) controller is a widely used type of controller used in applications ranging from a simple thermostat to complex system like aircraft. This presentation will explore the implementation of a PID controller on the longitudinal model of an aircraft, in this case, the Boeing 747. This model uses linearized equations of motion to describe the longitudinal motion of the aircraft using four states and two actuators. The presentation will discuss the modelling of the aircraft’s longitudinal dynamics, and the process of developing and tuning the PID controller to obtain a desirable response.
Faculty Advisor:
Professor Jonathan Rogers
Nathan Wong
Title:
Development of an LLM Assistant for Generation of DO-178C Compliant Unit Tests
Abstract:
A crucial aspect of aircraft safety and reliability is the testing, verification, and validation of avionics software, and the DO-178C document is the standard by with the FAA and EASA evaluate this software. With regards to testing, the DO-178C prescribes creating unit tests with 100% code coverage for any software that influences pilot workload or safety of the aircraft (design assurance rating of D or higher). However, this unit generation task is becoming very time consuming to do manually as the amount of software in modern aircraft grows. One such tool that has recently been used to generate unit tests in a more automated manner is large language models (LLMs). LLMs have been under intense development in recent years and have strongly impacted the way industries gather, analyze, and transform data, and one task they have been shown to be useful in is unit test generation. This presentation explores how to improve upon base LLMs to generate DO-178C compliant unit tests for avionics software, along with important considerations that must be made when using an LLM to do so.
Faculty Advisor:
Professor Olivia Pinon Fischer