Public Research Seminar by Advanced Materials Thrust, Function Hub, HKUST(GZ) - Machine learning accelerated thermodynamic modeling for materials design
Thermodynamics is an important foundation for materials design. However, thermodynamic modeling suffers high-dimensional challenges such as huge space of compositions and conditions and numerous configuration of atoms. In this presentation, I will first introduce the methods of thermodynamic modeling and their applications in materials desgin, and then talk about some recent progresses in acceleration of thermodynamic modeling by state-of-the-art machine learning techniques, including a deep learning framework of thermodynamic modeling and graph neural network accelerated Monte Carlo simulations
Dr. Pinwen Guan is currently a postdoctoral research associate in Sandia National Laboratories. He earned his B.S. degree in physics and Ph.D. degree in materials science from Tsinghua University and Pennsylvania State University respectively. He has focused on research and innovation in computational materials science, especially computational thermodynamics, and has been recently devoted to combining conventional materials modeling techniques with machine learning. His works have been applied in photovoltaics, superconductivity, energy storage, additive manufacturing and so on, leading to 24 research papers, 5 open-source codes and 4 US patents, with recognitions from renowned experts in the field.