☆ 4.0 Article

Segmentation of high-speed flow fields using physics-informed clustering

APPLICATIONS IN ENERGY AND COMBUSTION SCIENCE (2023)

Related references

Note: Only part of the references are listed.

Article Computer Science, Interdisciplinary Applications

Highly-scalable GPU-accelerated compressible reacting flow solver for modeling high-speed flows

Ral Bielawski et al.

Summary: Emerging supercomputing systems combine CPUs and GPUs to reach exascale capabilities with minimal energy footprint. This presents challenges for fluid solvers due to the differences in hardware architecture and operation between GPUs and CPUs. This work presents a general approach for efficient implementation of finite-volume based reacting flow solvers on heterogeneous systems. Specific algorithms are developed to handle compressible reacting flows, including chemical reactions, convection terms, and turbulence, ensuring GPU-based efficiency. The approach is demonstrated using the OpenFOAM software and shows excellent scalability on a large number of GPUs (> 3000) with constant throughput for a large number of control volumes.

COMPUTERS & FLUIDS (2023)

Add to Collection

Article Energy & Fuels

Data-driven discovery of heat release rate markers for premixed NH3/H2/air flames using physics-informed machine learning

Cheng Chi et al.

Summary: This study utilizes physics-informed machine learning to discover optimal HRR markers for premixed NH3/H-2/air flames. The identified markers show high reconstruction quality for various flame conditions and are not limited to specific kinetic mechanisms.

FUEL (2022)

Add to Collection

Article Thermodynamics

Numerical simulation of stable and unstable ram-mode operation of an axisymmetric ethylene-fueled inlet-isolator-combustor configuration

Caleb A. Hash et al.

Summary: Large-eddy simulations were conducted to investigate stable and unstable ramjet operational modes. The results showed that the trace levels of atomic oxygen in the experiment had a significant impact on the combustion process. A partially-premixed flame structure was observed during combustion with 1% atomic oxygen in the free stream, while a rich premixed flame was observed when the free stream was considered as pure air. The simulations of unstable ram-mode operation also indicated a sensitivity to the free-stream composition, and a reduction in atomic oxygen concentration yielded good agreement with experimental observations.

COMBUSTION AND FLAME (2022)

Add to Collection

Article Materials Science, Multidisciplinary

An approximate inertial manifold (AIM) based closure for turbulent flows

Maryam Akram et al.

AIP Advances (2022)

Add to Collection

Article Engineering, Aerospace

Large Eddy Simulation of Combustion for High-Speed Airbreathing Engines

Christer Fureby et al.

Summary: This paper reviews the fundamental models and methods required for LES of high-speed combustion, with a focus on combustion models and numerical simulation methods. Through experiments and comparisons, the results show good agreement.

AEROSPACE (2022)

Add to Collection

Article Thermodynamics

Data-driven Classification and Modeling of Combustion Regimes in Detonation Waves

Shivam Barwey et al.

Summary: This research implemented a data-driven approach to classify combustion regimes in detonation waves and demonstrated a procedure for domain-localized source term modeling based on these classifications. Through clustering analysis, distinctions between different combustion regimes within the detonation wave structure were illuminated. By using artificial neural networks, localized source term modeling was guided, leading to improved estimations of source terms.

FLOW TURBULENCE AND COMBUSTION (2021)

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 Mechanics

Using approximate inertial manifold approach to model turbulent non-premixed combustion

Maryam Akram et al.

Summary: The theory of inertial manifolds is used to develop reduced-order models of turbulent combustion by tracking system dynamics in a low-dimensional manifold without invoking laminar flame structures or statistical assumptions. The concept in approximate inertial manifolds is that slow dominant dynamical behavior is confined to a low-dimension manifold, while fast dynamics respond to the dynamics on the manifold instantaneously. This approach decomposes slow/fast dynamics and formulates an AIM by only exploiting the governing equations, showing promising results in capturing flame dynamics including extinction and reignition.

PHYSICS OF FLUIDS (2021)

Add to Collection

Article Mechanics

Mass flow rate effect on a rotating detonation combustor with an axial air injection

T. Sato et al.

Summary: By changing the flow rate of reactants injected into a rotating detonation combustor, interesting behavior of the system can be observed. A combination of experiments and numerical simulations shows that increasing mass flow rate can lead to a more uniform fuel-air mixture forming, resulting in a more ideal detonation effect with increased wave velocity.

SHOCK WAVES (2021)

Add to Collection

Article Energy & Fuels

A Neural Network-Inspired Matrix Formulation of Chemical Kinetics for Acceleration on GPUs

Shivam Barwey et al.

Summary: This study explores the use of GPU-accelerated neural networks to speed up high-fidelity simulations of turbulent flames with detailed chemical kinetics. By recasting Arrhenius kinetics as a neural network using matrix-based formulations, GPU-optimized linear algebra libraries can be utilized without the need for training. The performance analysis focuses on the scalability of source term calculations with reaction mechanism complexity, revealing trends in cell number effects on GPU saturation and speedup. The matrix-based method enables highly efficient GPU performance and near-peak performance in saturated regimes.

ENERGIES (2021)

Add to Collection

Article Thermodynamics

Adaptive chemistry via pre-partitioning of composition space and mechanism reduction

Giuseppe D'Alessio et al.

COMBUSTION AND FLAME (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders

Kookjin Lee et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

Add to Collection

Article Engineering, Mechanical

Data-based analysis of multimodal partial cavity shedding dynamics

Shivam Barwey et al.

EXPERIMENTS IN FLUIDS (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

A priori analysis of reduced description of dynamical systems using approximate inertial manifolds

Maryam Akram et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

M. Raissi et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2019)

Add to Collection

Article Thermodynamics

Experimental data-based reduced-order model for analysis and prediction of flame transition in gas turbine combustors

Shivam Barwey et al.

COMBUSTION THEORY AND MODELLING (2019)

Add to Collection

Article Computer Science, Artificial Intelligence

How much can k-means be improved by using better initialization and repeats?

Pasi Franti et al.

PATTERN RECOGNITION (2019)

Add to Collection

Article Engineering, Aerospace

Data-Driven One-Dimensional Modeling of Pseudoshocks

Romain Fievet et al.

JOURNAL OF PROPULSION AND POWER (2019)

Add to Collection

Article Thermodynamics

Emerging trends in numerical simulations of combustion systems

Venkat Raman et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2019)

Add to Collection

Article Thermodynamics

Principal component analysis coupled with nonlinear regression for chemistry reduction

Mohammad Rafi Malik et al.

COMBUSTION AND FLAME (2018)

Add to Collection

Article Mechanics

Spectral proper orthogonal decomposition and its relationship to dynamic mode decomposition and resolvent analysis

Aaron Towne et al.

JOURNAL OF FLUID MECHANICS (2018)

Add to Collection

Article Thermodynamics

Understanding scramjet combustion using LES of the HyShot II combustor

K. Nordin-Bates et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2017)

Add to Collection

Article Engineering, Aerospace

Wall-Modeled Large-Eddy Simulation of Autoignition-Dominated Supersonic Combustion

Graham V. Candler et al.

AIAA JOURNAL (2017)

Add to Collection

Article Thermodynamics

An efficient flamelet-based combustion model for compressible flows

Amirreza Saghafian et al.

COMBUSTION AND FLAME (2015)

Add to Collection

Article Thermodynamics

A pre-partitioned adaptive chemistry methodology for the efficient implementation of combustion chemistry in particle PDF methods

Youwen Liang et al.

COMBUSTION AND FLAME (2015)

Add to Collection

Article Thermodynamics

Reduced-order PCA models for chemical reacting flows

Benjamin J. Isaac et al.

COMBUSTION AND FLAME (2014)

Add to Collection

Article Mechanics

Cluster-based reduced-order modelling of a mixing layer

Eurika Kaiser et al.

JOURNAL OF FLUID MECHANICS (2014)

Add to Collection

Article Thermodynamics

MG-local-PCA method for reduced order combustion modeling

Axel Coussement et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2013)

Add to Collection

Article Thermodynamics

LES-based Eulerian PDF approach for the simulation of scramjet combustors

Heeseok Koo et al.

PROCEEDINGS OF THE COMBUSTION INSTITUTE (2013)

Add to Collection

Article Computer Science, Interdisciplinary Applications

A multivariate quadrature based moment method for LES based modeling of supersonic combustion

Pratik Donde et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2012)

Add to Collection

Article Mechanics

Dynamic mode decomposition of numerical and experimental data

Peter J. Schmid

JOURNAL OF FLUID MECHANICS (2010)

Add to Collection

Article Mechanics

Self-conditioned fields for large-eddy simulations of turbulent flows

Stephen B. Pope

JOURNAL OF FLUID MECHANICS (2010)

Add to Collection

Article Mathematics, Applied

Using adaptive proper orthogonal decomposition to solve the reaction-diffusion equation

Michael A. Singer et al.

APPLIED NUMERICAL MATHEMATICS (2009)

Add to Collection

Article Computer Science, Artificial Intelligence

Centroidal voronoi tessellation algorithms for image compression, segmentation, and multichannel restoration

Q Du et al.

JOURNAL OF MATHEMATICAL IMAGING AND VISION (2006)

Add to Collection

Article Thermodynamics

Hybrid large-eddy simulation/Lagrangian filtered-density-function approach for simulating turbulent combustion

V Raman et al.

COMBUSTION AND FLAME (2005)

Add to Collection

Article Computer Science, Interdisciplinary Applications

K-means clustering for optimal partitioning and dynamic load balancing of parallel hierarchical N-body simulations

YM Marzouk et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2005)

Add to Collection

Article Mathematics, Applied

On algorithms for permuting large entries to the diagonal of a sparse matrix

IS Duff et al.

SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS (2001)

Add to Collection

© Peeref 2019-2024. All rights reserved.