Search

13 Results

A Tailored Convolutional Neural Network for Nonlinear Manifold Learning of Computational Physics Data using Unstructured Spatial Discretizations
Post, January 10, 2022 • J. Tencer and K. Potter SIAM J. Sci. Comput., 43(4), A2581–A2613 We propose a nonlinear manifold learning technique based on deep convolutional autoencoders that is appropriate for model order reduction of physical systems in complex geometries. Convolutional neural networks have proven to be highly advantageous for compressing data arising from systems...

Efficient Sampling Methods for Machine Learning Error Models with application to Surrogates of Steady Hypersonic Flows
Post, March 10, 2022 • Elizabeth H. Krath, David S. Ching and Patrick J. Blonigan AIAA 2022-1249 This paper presents an investigation into sampling strategies for reducing the computational expense of creating error models for steady hypersonic flow surrogate models. The error model describes the quantity of interest error between a reduced-order model prediction and a full-order model...

Home
Page • About Our Group Our research focuses on methods to enable engineering design and analysis with intrusive and non-intrusive data-driven surrogate models. Simulating parameterized systems of equations is ubiquitous in science and engineering. It is often the case that solving such systems with high level of accuracy is a computationally intensive...

Model Reduction of Hypersonic Aerodynamics with Residual Minimization Techniques
Post, March 10, 2022 • David S. Ching, Patrick J. Blonigan, Francesco Rizzi and Jeffrey A. Fike AIAA 2022-1247 High-fidelity hypersonic aerodynamic simulations require extensive computational resources, hindering their usage in hypersonic vehicle design and uncertainty quantification. Projection-based reduced-order models (ROMs) are a computationally cheaper alternative to full-order simulations that can provide major speedup with marginal loss of accuracy...

Postdoctoral Appointee in Computational Science for Reduced Order and/or Surrogate Modeling
Post, March 9, 2022 • Location: Livermore, CAFull‐time/Temporary We are seeking a Postdoctoral Appointee at Sandia National Laboratories to join ourdepartment! This position involves research on surrogate and reduced order modelingwith an eye towards developing methods that can be deployed to large‐scalecomputational physics simulations. This position will also involve collaboration withdomain experts to apply these...

Projection-Based Model Reduction for Coupled Conduction-Enclosure Radiation Systems
Post, March 10, 2022 • Victor E Brunini, Eric Parish, John Tencer, and Francesco Rizzi ASME Journal of Heat Transfer A projection-based reduced order model (pROM) methodology has been developed for transient heat transfer problems involving coupled conduction and enclosure radiation. The approach was demonstrated on two test problems of varying complexity. The reduced order...

Software
Page • Project Website A minimally-intrusive interface that can be useful for a variety of purposesNumerous model reduction routines, including Galerkin, least-squares, and windowed least-squares projectionsSupport for arbitrary datatypes via generic programming and custom operationsBuilt-in support to use Eigen, Kokkos, and Trilinos (with more in progress)Several time integration schemes and nonlinear solverspressio4py:...