Jacqueline O'Connor

Professor, Department of Mechanical Engineering
Pennsylvania State University

Use of information networks for understanding hydrodynamic instability on complex wake flows

August 27, 2026

Hydrodynamic instability drives the performance and operability of a number of aerospace systems, including aircraft, rotorcraft, and propulsion systems. Identifying the source of the instability and its dynamics in complex configurations is challenging, particularly in highly turbulent, three-dimensional environments. In this talk, we'll explore the use of information network theory for identifying the critical regions of unstable flows. We apply these methods to wake flows, both single- and multi-wake systems, to explore opportunities for better insight into complex flow behaviors.

Biosketch

Dr. Jacqueline O'Connor is a professor of mechanical engineering at Penn State and directs the Reacting Flow Dynamics Laboratory. She and her students study issues related to combustor operability, alternative fuels, and high-temperature material durability and heat transfer for power and propulsion applications. She received a B.S. from MIT and an M.S. and Ph.D. from Georgia Tech. She was a post-doctoral researcher at Sandia National Laboratories before starting at Penn State in 2013. She is a fellow of the American Society of Mechanical Engineers and an associate fellow of the American Institute of Aeronautics and Astronautics.

Aditya Nair

Associate Professor, Department of Mechanical Engineering
University of Nevada, Reno

Data-driven approaches for simulation, modeling and control of unsteady fluid flows

September 3, 2026

Unsteady flows are expensive to simulate and hard to control. But their dynamics are rarely spread evenly. The physics that matters concentrates in a few regions of space and a few moments in time. This talk presents a suite of data-driven strategies built on that premise.

In space, dominant balance analysis locates where the governing physics is active. An adaptive mesh refinement framework uses this to allocate resolution where it is needed, cutting cost without sacrificing accuracy. Force and moment partitioning offers the complementary view, attributing unsteady loads to the specific vortical structures that generate them.

In time, spectral POD with triadic interaction mapping reveals how energy transfer between scales is regulated by modal amplitude. Phase-amplitude reductions identify when an oscillatory flow is most receptive to actuation. Cluster-based latent control coarse-grains the dynamics into transitions among representative states, and acts on those states directly in a learned latent space, with no model of the full system required.

Biosketch

Dr. Aditya G. Nair is an Associate Professor in the Department of Mechanical Engineering at the University of Nevada, Reno. His research interests are in the areas of computational fluid dynamics, high-performance computing, data science, and control theory focused on modeling and control of unsteady fluid flows. He received his M.S. from University of Michigan in 2013 and Ph.D. from Florida State University in 2018. Dr. Nair is the recipient of the Department of Energy Early Career Award in 2022, AFOSR DEPSCoR award in 2023 and is a founding member of the NSF AI institute of Dynamic Systems.

Matthew Krull

Doctoral Candidate, Department of Mechanical Engineering
Pennsylvania State University

Experimental and computational characterization of transonic turbine relevant flow fields

September 17, 2026

Simultaneous pressure and velocity measurements of compressible flows are essential for understanding fundamental flow phenomena such as shockwave–boundary layer interactions, influence of surface cooling/film cooling, or supersonic jet noise. To date, there are few measurement techniques that can obtain both quantities in the flow simultaneously. Of particular interest is development of non-intrusive, spatially resolved techniques since probes can interfere with the flow, and spatial resolution of complex flow interactions is important to identify sources of noise. Recent developments in particle image velocimetry have shown that by measuring the velocity field and its gradients, the pressure field can be inferred with adequate boundary information and assumptions about the flow (namely, either that it is incompressible, or is adiabatic if compressible).

This presentation will focus on establishing a robust algorithm to determine pressure from velocimetry measurements for adiabatic flows, and developing a framework to acquire temperature from thermographic velocimetry measurements for nonadiabatic flows. An algorithm to extract three-dimensional pressure from tomographic particle image velocimetry data for compressible nonadiabatic flows has been completed and demonstrated. Additionally, experimental measurements of the wake behind large-scale and true-scale airfoils have been measured, showing that pressure information can be experimentally obtained from particle image velocimetry data. Large Eddy Simulation (LES) has also been utilized to capture Reynolds- and Favre-averaged quantities to evaluate their impact on the pressure reconstruction algorithm.

Biosketch

Mr. Matthew Krull is a Ph.D student in Mechanical Engineering and a NASA Advanced Air Vehicles Program (AAVP) graduate fellow at Penn State University. He works in the Experimental and Computational Convection Laboratory (ExCCL) studying turbine aerodynamics and heat transfer. He received his M.S. in Mechanical Engineering at Penn State University, and his B.S. in Mechanical Engineering and a minor in Physics at Penn State Behrend.

Chaopeng Shen

Professor, Department of Civil Engineering
Pennsylvania State University

Sensitivity-aware neural operator learning: from Navier–Stokes rollouts to flood and tsunamis

September 24, 2026

Data-driven surrogates now approximate PDE solution maps orders of magnitude faster than solvers, yet the tasks we most need them for – parameter inversion, data assimilation, design and control, uncertainty quantification – consume the solver's sensitivities, not its outputs. Operators trained to fit fields alone reproduce flows while getting ∂u/∂p wrong, then fail at those tasks and drift in long rollouts. Enforcing sensitivity is what makes a fast surrogate usable. Matching the solver's Jacobian during training (SC-FNO, ICLR 2025) preserves field accuracy while enabling inversion over many parameters, with less data and shorter training, across equations and operator architectures. It also changes data scaling: with sensitivities enforced, information per sample grows with input dimensionality, blunting the curse of dimensionality for spatially distributed inputs and letting us reconstruct earthquake-induced seafloor deformation from sparse buoy records and forecast tsunami propagation with solver-level consistency in minutes. For autoregressive rollouts, refining each predicted window against the governing equation in its integral form, with the network frozen and no retraining, cuts error by 83% for turbulent Navier–Stokes, 40–72% for Burgers and 33–61% for the Richards equation of variably saturated flow. Because real domains are meshes and coastlines rather than periodic boxes, I will present spectral operators that learn on irregular geometries and transfer across them, and a multi-resolution shallow-water emulator that resolves wet–dry fronts while conserving volume under bathymetric uncertainty. Finally, gradient-faithful surrogates can stand in for solvers inside differentiable models, where neural closures are trained through the physics across thousands of catchments, recovering full-solver learning quality at roughly tenfold speedup.

Biosketch

Dr. Chaopeng Shen is Professor in Civil Engineering at The Pennsylvania State University. He received the Ph.D. degree in environmental engineering from Michigan State University, East Lansing, MI, USA, in 2009. His PhD research focused on computational hydrology and he developed the hydrologic model Process-based Adaptive Watershed Simulator (PAWS), which was later coupled to the community land model to study the interactions between hydrology and ecosystem. He was a Post-Doctoral Research Associate with the Lawrence Berkeley National Laboratory, Berkeley, CA, USA, from 2011 to 2012, working on high-performance computational geophysics. His later efforts focused on harnessing the big data and machine learning (ML) and physics-informed ML opportunities in advancing hydrologic predictions and understanding. As an early advocate for ML in geosciences, he has written technical, editorial, review and collective opinion papers on hydrologic deep learning to call to attention the emerging opportunities for scientific advances. He promotes differentiable modeling, which seamlessly integrates neural networks and physics for knowledge discovery and is now being implemented by NOAA as a candidate for the next-generational National Water Model. In addition, his research interests also include floodplain systems, scaling issues, process-based hydrologic modeling, and hydrologic data mining. He is currently Editor of Water Resources Research, Chief Editor for Frontiers in AI: Water and AI, and previously an inaugural Editor of Journal of Geophysical Research: Machine Learning and Computation.