Project Description

This project focuses on developing physics-informed AI digital twins that integrate crystal plasticity finite element (CPFE) simulations, additive manufacturing data, and in situ sensing to predict the microstructural and mechanical evolution of high-temperature structural alloys. By coupling physics-based modeling with machine learning, the research will enable real-time optimization of process parameters and component design, improving reliability, efficiency, and sustainability in advanced manufacturing and energy systems.

The digital-twin framework will incorporate adaptive learning to refine predictions as new data become available, creating a feedback loop between simulation, sensing, and design. This approach will allow materials and components to be virtually tested under diverse operating conditions, significantly reducing the time and cost required for physical prototyping and qualification. The resulting models and data infrastructure will support autonomous materials design and intelligent manufacturing workflows applicable to high-performance alloys in electric vehicles, robotics, and clean energy technologies.

Start Date

February 2, 2026

Postdoc Qualifications

- Ph.D. in Mechanical Engineering, Materials Science, or a related field
- Experience with finite element modeling, machine learning, and materials characterization
- Strong programming skills in Python, PyTorch, or TensorFlow
- Demonstrated expertise in constitutive modeling, microstructure-property relationships, or additive manufacturing
- Interest in AI integration, digital twins, and intelligent manufacturing systems 

Co-advisors

Prof. Karthik Ramani
Donald W. Feddersen Distinguished Professor of Mechanical Engineering
Purdue School of Mechanical Engineering
Email: ramani@purdue.edu
Website: https://engineering.purdue.edu/ME/People/ptProfile?resource_id=12331

Prof. David N. Johnson
Associate Professor of Materials Engineering
Purdue School of Materials Engineering
Email: davidjoh@purdue.edu
Website: https://engineering.purdue.edu/MSE/people/ptProfile?resource_id=11211 

Bibliography

Baweja, S., Messner, M.C. “Development of predictive model for accurate rupture time from multiaxial creep in Alloy 709 with physics-based simulations.” Computational Materials Science, 2024.

Baweja, S., Joshi, S.P. “Predicting textural variability effects in the anisotropic plasticity and stability of hexagonal metals: Application to magnesium and its alloys.” International Journal of Plasticity, 2020.

Baweja, S., et al. "Predicting long-term stress relaxation on Alloy 709 using the crystal plasticity finite element method". Argonne National Laboratory, 2024.

Winovich, N., Ramani, K., and Lin, G. “ConvPDE-UQ: Convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains.” Journal of Computational Physics, 2019.

Priya, P., Johnson, D.R., and Krane, M.J.M. “Precipitation during cooling of 7XXX aluminum alloys.” Computational Materials Science, 2017.