ASTRO Lab - Teaching

Courses

ME 4654/5794/5984: Optimization Techniques in Engineering (Spring 2019-2025)

Fundamental concepts for optimization, classification of optimization problems in engineering, optimality conditions, linear, gradient-based, and evolutionary optimization algorithms, the concept of inverse design, reliability and robustness-based optimization, and sensitivity analysis. Topics will include optimality criteria, Lagrangian problem formulation, Karush-Kuhn Tucker conditions, dual problem definition, step size calculation methods, linear programming, 1st and 2nd order gradient-based algorithms, evolutionary strategies for optimization, sensitivity analysis, and introduction to the concepts of reliability and robustness based optimization.

Learning outcomes:

Having successfully completed this course, the student will be able to:

• Fundamentally understand the mathematical concepts behind optimization and optimality.

• Define an engineering problem as a forward and/or inverse optimization problem.

• Understand the concepts behind linear, gradient-based and evolutionary optimization algorithms.

• Perform sensitivity analysis and integrate the sensitivity scheme to the optimization solution.

• Differentiate deterministic, reliability-based, and robustness-based optimization problems.

• Identify the most appropriate optimization technique to solve different engineering problems.

Grading Policy:

Homework Assignments: 40%

Midterm Exam: 20%

Project: 40%

ME 3614: Mechanical Design I (Spring 2018, Fall 2018-2022, Fall 2024)

Design of mechanical components subject to static and fatigue loads. Design using screws, fasteners, springs and bearings. Computer-aided design using transfer matrix and finite element methods. Pre: ESM 2204, MATH2114, MATH2204, and MATH2214.

Course outcomes:

• Estimate the stresses (or deflections) in a structural mechanical part

• Select an appropriate theory of failure applicable to the service of the structural mechanical part

• Apply static theories of failure and analyze (or design) structural mechanical parts

• Apply fatigue theories of failure to analyze or design structural mechanical parts

• Analyze and design power screws and fastener joints in static and fatigue service

• Analyze and design axial (extension and compression) and torsional springs in static and fatigue service

• Conduct a two-dimensional finite element analysis

ME 5984: Data-Driven Modeling and Materials Design (Spring 2025)

Fundamental concepts for data-driven modeling as well as state-of-the-art computational methods used for the design of different materials systems, including metals, metallic alloys, polymers, composites, and architected systems.

Learning outcomes:

Having successfully completed this course, the student will be able to:

• Learn the concept of data-driven modeling, with specific applications to material systems

• Learn the methodology behind widely used ML methods in materials modeling and design

• Identify appropriate data-driven and ML techniques to solve a prescribed modeling and design problem for a specific material class (metals, alloys, polymers, composites, architected materials)

• Understand fundamental concepts related to design and materials design

• Create alternative data-driven solution strategies to address prescribed materials modeling and design problems

Grading Policy:

In-Class Quizzes: 10%

Discretionary/Participation: 10%

In-Class Presentations: 40%

Final Project and Presentation: 40%

ME 4015: Senior Design Project (2019-2024)

Advisor of "Energy Efficiency Solutions for HVAC" sponsored by Alfa Laval (Fall 2023, Spring 2024)
Advisor of "NAVAIR Helicopter Tow Bar Redesign" sponsored by NAVAIR (Fall 2022, Spring 2023)
Advisor of "Fiber Removal from Paper Cones "Yarn Stripper" in collaboration with Universal Fibers (Fall 2020, Spring 2021)

Advisor of "Prosthetic Gym Hand" project in collaboration with Quality of Life Plus (Fall 2019, Spring 2020)

Undergraduate Research

Dr. Acar has been supervising research projects for undergraduate students at Virginia Tech, and offering summer research opportunities for visiting students.

Here are some examples.

Shape-to-Music (by Sofia Cassara and Buminhan Sansa, Fall 2024)

The Shape-to-Music project aims to encode metamaterial designs into unique musical representations. Each material is assigned a distinct sound based on the pixel values of its representative section images. The primary goal is to produce music that reflects material similarities - similar topologies generate similar melodies, while different topologies result in distinguishable tunes.

We approach this challenge using two methods:

Image-Based Approach: This method converts the pixel values of a material's image into musical notes, which are then combined to form its unique melody. Materials with similar pixel distributions - indicating similar topologies - produce comparable melodies.

Mechanical Modeling Approach: This method leverages the assumption that materials with similar topologies exhibit comparable mechanical properties. Simulations generate displacement and stress fields, which are then encoded into melodies using the Red-Green-Blue (RGB) values of these fields.

While both approaches create distinct sounds, we found that each effectively compares and differentiates metamaterial topologies through musical representation.

The Shape-to-Music project is supported by the National Science Foundation (NSF) CAREER Award # CMMI-2236947.

Physical Properties of a Composite Material (by Ben Anstrom, Spring 2019)

The study can be divided into two larger sections; first, using photographs of the composite cross section to approximate the volume fractions of the fifty different materials that make up the composite and then determining the material properties based off these volume fractions. Second, the volume fractions of the materials were to be manipulated to demonstrate the design versatility of the material. Yield stress and Young’s Modulus were the relevant design parameters for this study.

Using the available data on yield stress and stiffness, A GUI was created which shows the effect that the manipulation of certain volume fractions will have on the yield stress of the material.

Project Software

Microstructure Design for Implant Materials (by Melody Caloyannides, Spring 2019)

Many materials today contain properties that pose a risk to patients using the material as an implant. Risks include health issues as many of the common materials used have a high Young’s Modulus compared to that of the human bones’. Having a compatible Young’s Modulus is important to prevent bone resorption and achieve good bone remodeling. Traditional metals used for implants also have potential to be toxic with time for the human body. In this study, we model a non-toxic material, beta-Titanium alloy, which can also produce a Young's modulus value that is similar to the modulus of bones.

Melody presented the research in in Dennis Dean Undergraduate Research and Creative Scholarship Conference at Virginia Tech.

Design Optimization of Multi-Phase Materials (by Rick Catania, Abdalla Diraz, Dominic Maier, Armani Tagle, Fall 2019)

This work addresses various mathematical solution strategies adapted for design optimization of multiphase materials. The goal is to improve the structural performance by optimizing the distribution of multiple phases that constitute the material. Examples include the optimization of multiphase materials and composites with spatially varying fiber paths using a finite element analysis scheme.

We have published an invited journal paper in Mathematical Problems in Engineering [link]

Courses

ME 4654/5794/5984: Optimization Techniques in Engineering (Spring 2019-2025)

ME 3614: Mechanical Design I (Spring 2018, Fall 2018-2022, Fall 2024)

ME 5984: Data-Driven Modeling and Materials Design (Spring 2025)

ME 4015: Senior Design Project (2019-2024)

Undergraduate Research

Outreach Activities

* Materials Explorers

* Zonta International

* Collaboration with the Center of Engineering Diversity (CEED) at Virginia Tech.