Title:

Enabling Agile Materials Science through AI/ML

Speaker:

Dr. David Montes de Oca Zapiain

Affiliation:

Sandia National Laboratories

When:

Monday, November 14, 2022 at 2:00:00 PM   Add to Calendar

Where:

MRDC Building, Room 4211

Host:

Alexander Alexeev
alexander.alexeev@me.gatech.edu
4043853659

Abstract

The overall performance and behavior of a material is driven by spatially distributed features present in its internal structure. As a result, one needs to adequately account for the effect that the structure has on the performance in order to be able to provide realistic and accurate predictions of the material behavior. Nevertheless, this is not a trivial task because the internal structure of the material spans multiple length scales. Therefore, design efforts of components and materials need to be able to exchange information amongst the different length scales to adequately account for the effect the structure has on the overall performance. Unfortunately, the current process to characterize the effect the structure has at different length scales hampers the exchange of information amongst them. The main reason for that is that the current framework relies on time consuming and capital-intensive linkages that characterize the effect using expensive experimentation, characterization and high-fidelity simulations. As a result, the current process is not agile because the established linkages are not able to extract knowledge from previously performed experiments/simulations. This work showcases that these challenges can be surmounted by effectively integrating machine learning and artificial intelligence into the development of these linkages. The integration of AI and ML enables the development of data-driven models capable of extracting knowledge from previously performed experiments and simulations. As a result, these novel data-driven linkages use the extracted knowledge to provide accurate predictions of the effect that new/unseen structures will have at the desired length at a fraction of the cost. Thus, enabling a seamless flow of information amongst the different length scales.

This work was supported by the Laboratory Directed Research and Development program at Sandia National Laboratories. Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy National Nuclear Security Administration under contract DE-NA0003525. The views expressed in the article do not necessarily represent the views of the U.S. Department of Energy or the United States Government. Sand no. SAND2022-14963 A

Biography

David Montes de Oca Zapiain is originally from Mexico City Mexico and obtained his BS, MS and Ph.D. from the Woodruff School of Mechanical Engineering. He worked in Dr. Surya Kalidindi's research group and currently he is a senior member of the technical staff in the Department of Material and Data Science at Sandia National Laboratories. His research work focuses on developing accurate and computationally efficient process-structure and structure-property linkages using novel AI and machine learning techniques.

Notes

Refreshments will be served.