Zeliang Liu

TL;DR: A new data-driven computational framework is developed to assist in the design and modeling of new material systems and structures and includes the recently developed “self-consistent clustering analysis” method in order to build large databases suitable for machine learning.

TL;DR: A mechanistic, data-driven, two-scale approach is developed for predicting the behavior of general heterogeneous materials under irreversible processes such as inelastic deformation and is believed to open new avenues in parameter-free multi-scale modeling of complex materials, and perhaps in other fields that require homogenization of irreversible processes.

TL;DR: By discovering a proper topological representation of RVE with fewer degrees of freedom, this intelligent material model is believed to open new possibilities of high-fidelity efficient concurrent simulations for a large-scale heterogeneous structure.

TL;DR: In this article, the authors propose data-mining as an effective solution to understand the underlying physical mechanisms of additive manufacturing (AM) processes and material compositions, structures and properties in end-use products with arbitrary shapes.

TL;DR: In this article, a physics-informed neural network (PINN) framework was proposed to predict the temperature and melt pool dynamics during metal additive manufacturing (AM) processes with only a moderate amount of labeled data sets.