Grouped by year, newest first. See also Google Scholar.
arXiv
Non-intrusive operator inference is applied to DNS snapshots of a transitional flat-plate boundary layer, producing stable reduced models that capture Tollmien–Schlichting wave growth. Ut enim ad minim veniam.
Placeholder Journal of Fluid Mechanics, vol. 982, pp. A12
We present a wall-modelled large-eddy simulation framework for separated turbulent channel flow over periodic hills and show that a pressure-gradient-aware switch between equilibrium and non-equilibrium wall models improves reattachment prediction at modest cost. (Placeholder abstract.)
Example Journal of Multiphase Flow, vol. 161, no. 4, pp. 104455
Interface-resolved simulations of droplets in forced turbulence are used to map breakup regimes across Weber and Reynolds numbers. Lorem ipsum dolor sit amet, consectetur adipiscing elit.
Proceedings of the Placeholder Conference on Computational Engineering (PCCE), pp. 211-220
We train physics-informed neural networks on sparse RANS data to predict pressure and skin-friction distributions across a family of airfoils, cutting the cost of design sweeps by two orders of magnitude. Sed do eiusmod tempor incididunt ut labore.
Workshop on Placeholder High-Performance Computing (WPHPC), pp. 45-52
We describe a performance-portable finite-volume solver and report strong scaling to 512 GPUs on a canonical channel-flow benchmark. Duis aute irure dolor in reprehenderit in voluptate.
Placeholder Handbook of Computational Fluid Dynamics, 2nd ed., Example Press, pp. 301-330
An introductory chapter surveying machine-learning closures for RANS and LES, with worked examples and guidance on generalisation. Lorem ipsum dolor sit amet.
Example Institute of Technology
This thesis develops and assesses wall models for large-eddy simulation in flows with strong pressure gradients and separation. Excepteur sint occaecat cupidatat non proident.
No publications of this type yet.
PDFs are author preprints provided for personal use.