SRMP Seminar: June 12th

0 views
Skip to first unread message

AEM Communications

unread,
Jun 8, 2026, 12:30:26 PMJun 8
to AEM Regular Faculty, AEM Semester Teaching Faculty, AEM Research Contract Staff, AEM Seminar
Hi all,

We invite you to attend the upcoming SRMP seminar.

Friday, June 12, 2026
209 Akerman Hall
2:30pm-3:30pm

First Talk:
Title: MFINDER: A Multi-Fidelity, Physics-Informed Framework for Recovering and Denoising Incomplete Full-Field Kinematic Measurements

Abstract: Full-field Digital Image Correlation (DIC) measurements are central to data-driven constitutive modeling but are often corrupted by noise and missing data. We propose MFINDER (Multi-Fidelity Imputator and Denoiser), a physics-informed multi-fidelity framework for reconstructing smooth, complete displacement fields from noisy and incomplete DIC data of mechanical tests. MFINDER first calibrates a finite element (FE) model to available experimental measurements, producing a smooth low-fidelity field that reflects specimen geometry, loading, and boundary conditions. Gaussian Process Regression (GPR) then learns the discrepancy between sparse/noisy high-fidelity DIC data and the low-fidelity FE prediction, correcting FE-model bias while retaining physical smoothness and data efficiency. Additional constraints, including continuity, deformed specimen shape, and boundary conditions, further guide reconstruction. MFINDER is validated using synthetic DIC-like displacement fields from wedge indentation of a soft rectangular specimen and mode-I failure of a soft notched specimen. These cases include varying noise levels and spatially distributed missing data, especially near contact regions and crack/notch-induced strain concentrations. Compared with existing DIC imputation methods, MFINDER improves reconstruction accuracy by approximately an order of magnitude.

Speaker Bio: Nihar is a 2nd-year PhD student in the Soft Materials Mechanics lab, advised by Prof. Kshitiz Upadhyay. His motivation to pursue a Ph.D. stems from his curiosity about the complex biomechanics of soft tissues like the heart and the brain. His current research focuses on developing data-driven constitutive modeling techniques to understand and capture the response of brain tissue during traumatic brain injury (TBI). Besides his research in solid mechanics, he likes to sing and play the harmonium, and is trained in Indian Classical Music.

Second Talk:
Title: Numerical Simulation of Hypersonic Boundary Layers

Abstract: Hypersonic vehicles experience extreme heat flux and drag, which can be dramatically intensified by turbulence and vortices transporting hot air from the shock towards the vehicle surface. Accurately quantifying the relevant flow field structures in the boundary layer and their contributions to the heating and drag is critical to understanding performance and enabling design studies of hypersonic vehicles. This talk will provide a high-level overview of computational fluid dynamics (CFD) simulations of several hypersonic ground tests and flight tests, including HiFire-1, HiFire-5, BOLT II, and Oberkampf. This talk will discuss the physics of laminar-turbulent boundary layer transition, as well as new tools being developed to reduce computational costs and improve accessibility for designers.

Speaker Bio: Carter Vu is a fourth year PhD student with Dr. Graham Candler. His work focuses on high-speed boundary layer instability and transition to turbulence, particularly on the BOLT II flight test. He has completed internships at Aerojet Rocketdyne and the Johns Hopkins Applied Physics Laboratory. In addition to his conference publications, Carter has given invited talks at Kennedy Space Center through the Astronaut Scholarship, as well as ONERA, the French Aerospace Laboratory, through a NATO collaboration.

--
AEM Marketing and Communications 
Aerospace Engineering & Mechanics | aem.umn.edu
College of Science & Engineering | cse.umn.edu 
University of Minnesota Twin Cities | umn.edu
110 Union Street SE, Minneapolis, MN 55455
Reply all
Reply to author
Forward
0 new messages