Paper Authors

Abstract

NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract

Algorithmic Thinking and Matlab in a Computational Materials Science Course

Abstract

A course was developed to teach aspects of materials science, numerical methods, and programming in an integrated fashion. During the second teaching of the course, it was modified to enhance its delivery by focusing on the aspects which gave the students the most difficulty in its first offering: syntax and organization of operations in programming. This was achieved through the use of Matlab as a meta-language platform, development of Matlab tutorials for the course, and an emphasis on algorithmic thinking.

In this paper, algorithmic thinking involves developing a complete understanding of the operations required via hand calculations and block diagrams before attempting to generate any code. Students were graded on their ability to relate what the program/algorithm should do next verbally and pictorially and then tasked with translating those known operations into Matlab code using Matlab’s extensive help menus. The help menus allow users to employ keyword searches to find descriptions and examples of commands with the needed functionality.

Results of student projects show improvement from the first to second years. Student response to the course also shows an increased respect for Matlab as a useful engineering tool. In both years, students who were unable to verbally describe the needed operations in the programs generated less efficient or inoperable code.

Introduction

Computational Materials Science (CMS) is a cross-listed senior elective and graduate course in Mechanical Engineering that meets for 75 minutes twice weekly. The course is also part of a newly created Materials Science Concentration. The course covers topics in three fundamental areas: numerical techniques, geometric and potential-energy aspects of materials science, and programming. Numerical techniques primarily involve minimization of potential functions while the materials science concepts center around Lennard-Jones potential functions and the network structure of polymers. The course is offered biannually and was last offered in the Fall 2004-2005 academic year. During this last teaching, the course was revised to better facilitate student comprehension and application of concepts based on observations from previous teachings. The content of the 3 semester hour course is broken into roughly 40% numerical techniques, 25% materials science, and 35% programming using the text: Andrew Leach, Molecular Modeling: Principles and Applications3.

While students in mechanical engineering are well versed and comfortable with the theory behind numerical minimization techniques ("if df/dx=0 then you are done") and the basic theories behind materials science concepts ("the structure wants to achieve a minimum energy"), they are quite uncomfortable with programming. Furthermore, they rarely, at the undergraduate

• Citation

• Format
  • APA
  • APA - LaTeX bibitem
  • MLA
  • MLA - LaTeX bibitem
  • Bibtex
  • EndNote - RIS

von Lockette, P. (2006, June), Algorithmic Thinking And Matlab In Computational Materials Science Paper presented at 2006 Annual Conference & Exposition, Chicago, Illinois. 10.18260/1-2--725