☆ 4.6 Article

Applying physics-informed enhanced super-resolution generative adversarial networks to turbulent premixed combustion and engine-like flame kernel direct numerical simulation data

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2023)

Related references

Note: Only part of the references are listed.

Proceedings Paper Computer Science, Artificial Intelligence

AI super-resolution subfilter modeling for multi-physics flows

Mathis Bode

Summary: This study proposes a novel subfilter modeling approach based on AI super-resolution, which accurately models the small scales in complex applications using a physics-informed enhanced super-resolution generative adversarial network. The research demonstrates how this approach can accelerate complex simulations and provides examples in three fluid dynamics simulation setups. Moreover, the advantages and shortcomings of the proposed approach are discussed.

PROCEEDINGS OF THE PLATFORM FOR ADVANCED SCIENTIFIC COMPUTING CONFERENCE, PASC 2023 (2023)

Add to Collection

Article Thermodynamics

Using physics-informed enhanced super-resolution generative adversarial networks for subfilter modeling in turbulent reactive flows

Mathis Bode et al.

Summary: Turbulence poses challenges in predicting reactive flows accurately. The development of new turbulence closures is essential for combustion problems. Data-driven modeling has gained popularity in various fields, but applying deep neural networks in fluid dynamics remains challenging. The study introduces a novel subfilter modeling approach based on a generative adversarial network, improving generalization capability through a two-step training method.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2021)

Add to Collection

Article Biology

Pandemic drugs at pandemic speed: infrastructure for accelerating COVID-19 drug discovery with hybrid machine learning- and physics-based simulations on high-performance computers

Agastya P. Bhati et al.

Summary: The existing drug discovery process is costly, inefficient, and slow, but new opportunities lie at the interface between machine learning and physics-based methods. An innovative infrastructure development combines these methods to accelerate drug discovery, relying on supercomputing for high throughput. This approach has demonstrated viability for identifying lead antiviral compounds through repurposing on various supercomputers.

INTERFACE FOCUS (2021)

Add to Collection

Article Mechanics

Reynolds number scaling of burning rates in spherical turbulent premixed flames

Tejas Kulkarni et al.

Summary: In the flamelet regime of turbulent premixed combustion, the burning rates are primarily enhanced by surface wrinkling. Through direct numerical simulations of spherical turbulent premixed methane/air flames, the study reveals that the temporal evolution of the flame surface is influenced by Reynolds number while keeping the Karlovitz number constant. The brush thickness scales with the integral scale of the flow, while the wrinkling scale defined as the inverse of the peak surface density scales with Reynolds number and decreases as turbulence decays, leading to an increase in the area ratio and burning rate.

JOURNAL OF FLUID MECHANICS (2021)

Add to Collection

Article Thermodynamics

DNS study of the global heat release rate during early flame kernel development under engine conditions

Tobias Falkenstein et al.

COMBUSTION AND FLAME (2020)

Add to Collection

Article Thermodynamics

The role of differential diffusion during early flame kernel development under engine conditions - part II: Effect of flame structure and geometry

Tobias Falkenstein et al.

COMBUSTION AND FLAME (2020)

Add to Collection

Article Thermodynamics

The role of differential diffusion during early flame kernel development under engine conditions - part I: Analysis of the heat-release-rate response

Tobias Falkenstein et al.

COMBUSTION AND FLAME (2020)

Add to Collection

Article Thermodynamics

Role of large scale flow features on cycle-to-cycle variations of spark-ignited flame-initiation and its transition to turbulent combustion

Wei Zeng et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2019)

Add to Collection

Article Thermodynamics

Adaptive chemistry lookup tables for combustion simulations using optimal B-spline interpolants

Mathis Bode et al.

COMBUSTION THEORY AND MODELLING (2019)

Add to Collection

Article Physics, Fluids & Plasmas

Synthetic turbulent inflow generator using machine learning

Kai Fukami et al.

PHYSICAL REVIEW FLUIDS (2019)

Add to Collection

Article Multidisciplinary Sciences

Grandmaster level in StarCraft II using multi-agent reinforcement learning

Oriol Vinyals et al.

NATURE (2019)

Add to Collection

Article Thermodynamics

Empirical investigation of spark-ignited flame-initiation cycle-to-cycle variability in a homogeneous charge reciprocating engine

Philipp Schiffmann et al.

INTERNATIONAL JOURNAL OF ENGINE RESEARCH (2018)

Add to Collection

Article Mechanics

Geometry and scaling laws of excursion and iso-sets of enstrophy and dissipation in isotropic turbulence

Jose Hugo Elsas et al.

JOURNAL OF TURBULENCE (2018)

Add to Collection

Article Thermodynamics

On the extinction of igniting kernels in near-isotropic turbulence

Harshavardhana A. Uranakara et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2017)

Add to Collection

Article Thermodynamics

Optimal artificial neural networks and tabulation methods for chemistry representation in LES of a bluff-body swirl-stabilized flame

Matthias Ihme et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2009)

Add to Collection

Article Mechanics

Effects of initial radius on the propagation of premixed flame kernels in a turbulent environment

M. Klein et al.

PHYSICS OF FLUIDS (2006)

Add to Collection

© Peeref 2019-2024. All rights reserved.