Machine Learning–Enhanced Multiscale Modeling of Spatially Tailored Materials

Sponsor: U.S. National Science Foundation

Project No.: CMMI-2104383

Duration: Aug 15, 2021 – Aug 14, 2024

Principal Investigator: Professor Shaoping Xiao

Graduate Students: Siamak Attarian and Arunabha Batabyal

Summary

Spatially tailored materials, based on the concept of functionally graded materials, are composites consisting of two or more different materials. The volume fraction of each material continuously changes in space. These composites offer advantages over traditional ones due to the ability to use the predominant characteristics of the materials that make up the composites and tailor them to the conditions and environment in which they are operating. On the other hand, multiscale modeling is the use of models at different levels of resolution (in this case, the nanoscale, microscale, and macroscale) to describe a physical system. This award supports the development of an innovative multiscale method, enhanced by machine learning, to investigate the mechanics of spatially tailored materials under mechanical and thermal loading. The proposed method, if developed and validated by experimental testing, will accelerate the design of the next generation of composites for use in the automotive, aerospace, and biomedical industries. In addition, the award will support: (1) undergraduate teaching and learning in data science and engineering; (2) recruitment of female, underrepresented minority, and lesbian/gay/bisexual/transgender/queer students; and 3) outreach to K-12 students through university programs.

Modular deep reinforcement learning for continuous motion planning with temporal logic
M Cai, M Hasanbeig, S Xiao, A Abate, Z Kan
IEEE Robotics and Automation Letters 6 (4), 7973-7980, 2021

Machine learning in multiscale modeling of spatially tailored materials with microstructure uncertainties
S Xiao, P Deierling, S Attarian, A El Tuhami
Computers & Structures 249, 106511, 2021

Development of a 2NN-MEAM potential for boron
S Attarian, S Xiao
Journal of Micromechanics and Molecular Physics 5 (03), 2050008, 2020

A machine-learning-enhanced hierarchical multiscale method for bridging from molecular dynamics to continua
S Xiao, R Hu, Z Li, S Attarian, KM Björk, A Lendasse
Neural Computing and Applications 32 (18), 14359-14373, 2020

Investigation of the Observed Rupture Lines in Abdominal Aortic Aneurysms Using Crack Propagation Simulations
S Attarian, S Xiao, TC Chung, ES da Silva, ML Raghavan
Journal of biomechanical engineering 141 (7), 2019

Peridynamics with Corrected Boundary Conditions and Its Implementation in Multiscale Modeling of Rolling Contact Fatigue
MA Ghaffari, Y Gong, S Attarian, S Xiao
Journal of Multiscale Modelling 10 (01), 1841003, 2019

Atomistic simulation of diffusion bonding of dissimilar materials undergoing ultrasonic welding
A Samanta, S Xiao, N Shen, J Li, H Ding
The International Journal of Advanced Manufacturing Technology, 2019

Multiscale modeling and simulation of rolling contact fatigue
MA Ghaffari, Y Zhang, S Xiao
International Journal of Fatigue 108, 9-17, 2018