Machine learning paper appears on Patterns

Our group collaborated with Dr. Fei Zhou and other scientists at LLNL in utilizing machine learning to simulate materials microstructure evolution. This work is just published in the open-access journal Patterns from Cell Press:  https://doi.org/10.1016/j.patter.2021.100243

Material microstructure plays a key role in the processing-structure-property relationship of engineering materials. Microstructure evolution is commonly simulated by continuum models based on partial differential equations. We apply convolution recurrent neural networks to learn and predict several microstructure evolution phenomena of different complexities. The method is significantly faster than the traditional approach and capable of predicting the evolution process in systems with unknown material parameters. It provides a useful data-driven alternative to microstructure simulation.

Congratulations to the current and past group members (Kaiqi, Henry, Youtian and Shaoxun) who contributed to this work.