Partha Pratim Das, Ph.D.
Mechanical Engineer • Composite Materials • Structural Health Monitoring • Physics-Informed AI
"An artist cannot fail; it is a success to be one"
— Charles Horton CooleyI am currently working as a Research Scientist II at the University of Texas at Arlington Research Institute (UTARI), actively contributing to advanced composite mechanics and structural health monitoring research funded by the US Air Force Research Laboratory (AFRL).
I pursued my Ph.D. in Mechanical Engineering at the University of Texas at Arlington (UTA). During my graduate studies, I worked as a Graduate Research Assistant and Intern at the Institute for Predictive Performance Methodologies (IPPM) at UTARI, focusing on multiscale-multiphysics moisture degradation, dielectric state variable tracking, and scientific machine learning (PIML/SciML) for polymer composites.
"I think I am not unique. It's not necessary to be unique to be a useful one in this society — I believe. In short, I like to do what I love to do, I love to do what I can do... I am always craving for learning something new. I love my undergrad university — BUET's campus; I still cherish every moment I had passed there, in the classes or around the cafeteria, everywhere!"
Winner: ASC 4-Minute Doctoral Research Impact Competition & ASC PhD Research Scholarship
Recognized for doctoral breakthroughs in multimodal moisture degradation and AI-based composite prognostics.
Featured Research
Core scientific programs, doctoral thrusts, and specialized rigs
Multimodal Hygrothermal Aging & Dielectric Prognostics
Non-Fickian hindered diffusion (HDM) coupled with Maxwell’s electromagnetic equations and Physics-Informed Neural Networks (PINNs) to track moisture sorption kinetics and mechanical property degradation in GFRP composites.
AI-Assisted Prognostic Health Monitoring (PHM)
Artificial neural network (ANN) framework incorporating real-time dielectric state variables to predict residual strength and remaining useful life (RUL) under fatigue loading without prior run-to-failure historical records.
In-Situ 2D DIC (RealPi2dDIC) & Microtensile Testing
Open-source Raspberry Pi based 2D digital image correlation (RealPi2dDIC) providing real-time full-field sub-pixel strain mapping (<$100 setup), paired with high-precision in-situ microtensile testing apparatus.