Showing 1–2 of 2 results for author: Towara, M
-
Reducing Spatial Discretization Error on Coarse CFD Simulations Using an OpenFOAM-Embedded Deep Learning Framework
Authors:
Jesus Gonzalez-Sieiro,
David Pardo,
Vincenzo Nava,
Victor M. Calo,
Markus Towara
Abstract:
We propose a method for reducing the spatial discretization error of coarse computational fluid dynamics (CFD) problems by enhancing the quality of low-resolution simulations using deep learning. We feed the model with fine-grid data after projecting it to the coarse-grid discretization. We substitute the default differencing scheme for the convection term by a feed-forward neural network that int…
▽ More
We propose a method for reducing the spatial discretization error of coarse computational fluid dynamics (CFD) problems by enhancing the quality of low-resolution simulations using deep learning. We feed the model with fine-grid data after projecting it to the coarse-grid discretization. We substitute the default differencing scheme for the convection term by a feed-forward neural network that interpolates velocities from cell centers to face values to produce velocities that approximate the down-sampled fine-grid data well. The deep learning framework incorporates the open-source CFD code OpenFOAM, resulting in an end-to-end differentiable model. We automatically differentiate the CFD physics using a discrete adjoint code version. We present a fast communication method between TensorFlow (Python) and OpenFOAM (c++) that accelerates the training process. We applied the model to the flow past a square cylinder problem, reducing the error from 120% to 25% in the velocity for simulations inside the training distribution compared to the traditional solver using an x8 coarser mesh. For simulations outside the training distribution, the error reduction in the velocities was about 50%. The training is affordable in terms of time and data samples since the architecture exploits the local features of the physics.
△ Less
Submitted 5 September, 2024; v1 submitted 12 May, 2024;
originally announced May 2024.
-
Elimination Techniques for Algorithmic Differentiation Revisited
Authors:
Uwe Naumann,
Erik Schneidereit,
Simon Maertens,
Markus Towara
Abstract:
All known elimination techniques for (first-order) algorithmic differentiation (AD) rely on Jacobians to be given for a set of relevant elemental functions. Realistically, elemental tangents and adjoints are given instead. They can be obtained by applying software tools for AD to the parts of a given modular numerical simulation. The novel generalized face elimination rule proposed in this article…
▽ More
All known elimination techniques for (first-order) algorithmic differentiation (AD) rely on Jacobians to be given for a set of relevant elemental functions. Realistically, elemental tangents and adjoints are given instead. They can be obtained by applying software tools for AD to the parts of a given modular numerical simulation. The novel generalized face elimination rule proposed in this article facilitates the rigorous exploitation of associativity of the chain rule of differentiation at arbitrary levels of granularity ranging from elemental scalar (state of the art) to multivariate vector functions with given elemental tangents and adjoints. The implied combinatorial Generalized Face Elimination problem asks for a face elimination sequence of minimal computational cost. Simple branch and bound and greedy heuristic methods are employed as a baseline for further research into more powerful algorithms motivated by promising first test results. The latter can be reproduced with the help of an open-source reference implementation.
△ Less
Submitted 20 March, 2023;
originally announced March 2023.
