"When we look at materials now, we usually tune mechanical properties in one direction," Razavi explained. "For example, they can absorb shock in the 'x' direction, but they don't account for the 'y' or 'z' direction. Strengthening one direction might compromise properties in others."
With the support of a $313,087 grant from the National Science Foundation, the team will develop a deep-learning model based on physical laws to optimize the microarchitecture of composite materials. Liu added, "Imagine mixing two types of materials, like a solid stone and a soft gel. How can you design the distribution of these materials? They will have different mechanical properties depending on the direction."
The researchers plan to train their machine learning algorithms using thousands of computational models. Once the most promising designs are identified, Associate Professor Yanyu Chen from the University of Louisville will test them through 3D printing, X-ray imaging, and stress testing.
The inspiration for the project comes from Razavi's work on the human brain, specifically how brain folds form as grey matter grows over white matter. Brain tissue, with its different fiber tracts, has varying mechanical properties depending on direction-similar to the behavior Razavi and Liu aim to replicate in composite materials.
The team believes this research could lead to new, customized materials with tailored properties for a variety of uses, including lighter structures, advanced shock absorbers, and aerospace components. Liu noted, "This could be applied to everyday products too, like designing more comfortable shoes with customized mechanical properties."
Related Links
Binghamton University
Space Technology News - Applications and Research
| Subscribe Free To Our Daily Newsletters |
| Subscribe Free To Our Daily Newsletters |
