Announcements

First-Generation College Celebration 2024

The annual First-Generation College Celebration takes place on November 8th. We are looking for faculty/staff who are interested in collaborating on this year's celebration programming! If you or your office/area are interested in helping to plan programming; or if you are a first-generation college graduate that is interested in participating, but not planning to program please reach out to Maggie Weiss in the Center for Student Involvement - [email protected] -2925

Posted By: weissm

Karimi Thesis Defense

Sara Karimi Thesis Defense "Determination of Game-based Design Equilibria by using Machine Learning Approach" MS/Mechanical

July 29, 9am BMH 106

ABSTRACT
This thesis presents a comprehensive study on the application of artificial intelligence and
machine learning to enhance efficiency and precision in game-based design problems, with
specific focus on pressure vessels, bilevel problems with three followers, and speed reducers as
numerical examples. An AI-enhanced machine learning framework that optimizes complex
engineering designs beyond traditional methods has been proposed. The novel approach is
demonstrated through three example problems, each viewed as a game with set players, presenting
unique challenges and design requirements. The process begins by developing datasets from
specific problem intervals and features, using MATLAB tools to achieve optimized solutions.
These optimized results then serve as training data for a neural network, designed to predict
rational reaction sets of players involved in the design process, thereby facilitating more informed
and accurate decision-making. Simulation of the optimization problems is conducted to generate
comprehensive datasets for each example. In the pressure vessel problem, simulations were
performed to determine the optimal thickness, radius, and length values, resulting in minimized
weight while adhering to safety constraints. For the bilevel problem with three followers, the
simulation results provided data to train a neural network that effectively predicts rational reaction
sets, leading to improved optimization outcomes. In the speed reducer problem, the AI-enhanced
approach facilitated the prediction of optimal design parameters through the integration of neural
networks, demonstrating the practical application and efficiency of the proposed framework. By
integrating advanced machine learning techniques and formulating problems through game theory,
this approach simplifies computations and enhances the reliability and flexibility of engineering
solutions, providing a new era in efficient and robust design optimization.

Abstract

Posted By: browncr