Data-Driven Design of Tough 3D Printed Structures

SPRING 2018 RESEARCH INCUBATION AWARDEES

Pl: Emily Whiting, Computer Science, CAS
Co-Pls: Keith Brown, Mechanical Engineering, ENG; Elise Morgan, Mechanical Engineering, ENG

The project converges machine learning, physical experimentation, and design to address the general question of how to optimize a design when the fitness function can only be reliably determined through physical experimentation. Using a data-driven approach, it will combine low throughput physical experimentation with high throughput design to optimize and understand the failure of 3D printed parts. Designing structural parts that need to be resistant to failure requires an understanding of how the printing process affects susceptibility to failure. However, using traditional approaches, such understanding requires a detailed study of each printing material and process. The success of the proposed research program will overcome this challenge by providing insight on how to design tough materials and moving additive manufacturing into the realm of high-performance structural materials.

This work is funded by a Research Incubation Award made in January, 2018.