Tobias Horstmann
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Tobias Horstmann has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
Arthur Stück
1 shared publication- Ben Picard
Resource and Energy-efficient Engineering (TREE)
1 shared publication Denis Ricot
Renault Technoctr, 1 Ave Golf, F-78280 Guyancourt, France
1 shared publication- Dirk Reith
Hochschule Bonn-Rhein-Sieg
1 shared publication - Emmanuel Lévêque
CNRS
1 shared publication Gregorio Gerardo Spinelli
1 shared publication
Latest Publications
- Open Access
- 12 November 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (5) , 055308
Machine-learning-enhanced collision operator for the lattice Boltzmann method based on invariant networks
Integrating machine learning () techniques in established numerical solvers represents a modern approach to enhance computational fluid dynamics simulations. Within the lattice Boltzmann method (LBM), the collision operator serves as an ideal entry point to incorporate machine learning () techniques to enhance its accuracy and stability. In this work, an invariant neural network is constructed, acting on an equivariant collision operator, optimizing the relaxation rates of nonphysical moments. This optimization enhances robustness to symmetry transformations and ensures consistent behavior across geometric operations. The proposed neural collision operator () is trained using forced isotropic turbulence simulations driven by spectral forcing, ensuring stable turbulence statistics. The desired performance is achieved by minimizing the energy spectrum discrepancy between direct numerical simulations and underresolved simulations over a specified wave number range. The loss function is further extended to tailor numerical dissipation at high wave numbers, ensuring robustness without compromising accuracy at low and intermediate wave numbers. The 's performance is demonstrated using three-dimensional Taylor-Green vortex (TGV) flows, where it accurately predicts the dynamics even in highly underresolved simulations. Compared to other LBM models, such as the and operators, the exhibits improved accuracy while maintaining stability. In addition, the operator shows robust performance in alternative configurations, including turbulent three-dimensional cylinder flow. Finally, an alternative training procedure using time-dependent quantities is introduced. It is based on a reduced TGV model along with newly proposed symmetry boundary conditions. The reduction in memory consumption enables training at significantly higher Reynolds numbers, successfully leading to stable yet accurate simulations.
- 12 November 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (5) , 055308
Open AccessMachine-learning-enhanced collision operator for the lattice Boltzmann method based on invariant networks
Integrating machine learning () techniques in established numerical solvers represents a modern approach to enhance computational fluid dynamics simulations. Within the lattice Boltzmann method (LBM), the collision operator serves as an ideal entry point to incorporate machine learning () techniques to enhance its accuracy and stability. In this work, an invariant neural network is constructed, acting on an equivariant collision operator, optimizing the relaxation rates of nonphysical moments. This optimization enhances robustness to symmetry transformations and ensures consistent behavior across geometric operations. The proposed neural collision operator () is trained using forced isotropic turbulence simulations driven by spectral forcing, ensuring stable turbulence statistics. The desired performance is achieved by minimizing the energy spectrum discrepancy between direct numerical simulations and underresolved simulations over a specified wave number range. The loss function is further extended to tailor numerical dissipation at high wave numbers, ensuring robustness without compromising accuracy at low and intermediate wave numbers. The 's performance is demonstrated using three-dimensional Taylor-Green vortex (TGV) flows, where it accurately predicts the dynamics even in highly underresolved simulations. Compared to other LBM models, such as the and operators, the exhibits improved accuracy while maintaining stability. In addition, the operator shows robust performance in alternative configurations, including turbulent three-dimensional cylinder flow. Finally, an alternative training procedure using time-dependent quantities is introduced. It is based on a reduced TGV model along with newly proposed symmetry boundary conditions. The reduction in memory consumption enables training at significantly higher Reynolds numbers, successfully leading to stable yet accurate simulations.
- Open Access
- 17 February 2023
- Published by Elsevier in Computers & Fluids
- Vol. 255
HPC performance study of different collision models using the Lattice Boltzmann solver Musubi
- 17 February 2023
- Published by Elsevier in Computers & Fluids
- Vol. 255
Open AccessHPC performance study of different collision models using the Lattice Boltzmann solver Musubi
- 15 April 2022
- Published by Elsevier in Journal of Computational Physics
- Vol. 462
Consistent time-step optimization in the lattice Boltzmann method
- 15 April 2022
- Published by Elsevier in Journal of Computational Physics
- Vol. 462
Consistent time-step optimization in the lattice Boltzmann method