Research

Reduced Order Modeling

I am interested in constructing projection-based reduced order models (pROMs) of complex engineered systems. Specifically, I focus on pROMs for heat transfer, fluids, and ablation physics. I contribute to the Pressio project which enables rapid deployment of pROM technology to a large number of application codes. I am currently working on nonlinear manifold learning methods and sampling methods for least squares regression problems arising from pROMs.

Computational Heat Transfer

I have a broad range of interests in the area of computational heat transfer, including the simulation of complex physical systems in extreme environments and numerical methods for radiation transport, multiphase flow, and ablation systems.

Uncertainty Quantification

I am the lead developer for the PyNetUQ project for uncertainty propagation in large-scale networks. Each node in the network represents a potentially complex multiphysics simulation (or a surrogate model thereof). The advantages of decomposing models in this way are directly tying component/subassembly level validation data to system level model predictions, reducing model development time by enabling parallel development of component/subassembly models, and improving final design quality by allowing UQ to be incorporated earlier in the design process.