 |
Center for Nonlinear Studies |
The discovery and design of new materials relies on understanding the connections between processing, structure, and performance. Through probabilistic methods, we study the connection between processing and structure, specifically on lengths scales between microns and millimeters. Oftentimes this microstructure of a material will lead to surprising effects: magnificent colors or super-elasticity for example.
Starting with a collection of data in form of images generated by phase-field type models, we develop a metric based on the correlation functions. Equipped with a measure of similarity, we can apply a plethora of information-theoretic and machine-learning techniques in order to understand better and make predictions on the connection between processing and the resulting structure.