DLBFoam: An open-source dynamic load balancing model for fast reacting flow simulations in OpenFOAM

Tekgül, Bulut; Peltonen, Petteri; Kahila, Heikki; Kaario, Ossi; Vuorinen, Ville

doi:10.1016/j.cpc.2021.108073

Cited by 49 publications

(42 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome the computational load imbalance among processors, an open-source dynamic load balancing model (DLBFoam), developed recently by Tekgül et al (2021), is utilized to distribute the computation of chemistry problems equally among all processors using the MPI communication protocol. The DLBFoam model may be the first open-source implementation of chemistry dynamic load balancing reported in the literature enabling such computationally expensive simulations.…”

Section: Reactive Flow Solvermentioning

confidence: 99%

“…2. The interested reader is referred to Zirwes et al (2020) for detailed information on the cubic discretization scheme implemented in OpenFOAM, and Tekgül et al (2021) and Morev et al (2022) for comprehensive explanation about the customized solver employed here.…”

Section: Reactive Flow Solvermentioning

confidence: 99%

“…It should be emphasized that, to the best of our knowledge, performing such computationally expensive simulations were not feasible with the currently available open-source solvers until recently. In this work, the recently developed open-source dynamic load balancing model (DLBFoam) (Tekgül et al 2021) is utilized for simulating these test conditions in the open-source CFD toolbox OpenFOAM. Since the majority of works in the literature utilize methane/air flames in their studies, it is of particular interest to use it as a benchmark fuel/air mixture to compare its results with ammonia/air flames.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Inner Flame Front Structures and Burning Velocities of Premixed Turbulent Planar Ammonia/Air and Methane/Air Flames

Tamadonfar

Karimkashi

Kaario

et al. 2022

Flow Turbulence Combust

Self Cite

View full text Add to dashboard Cite

Ammonia ($$\mathrm {NH_3}$$ NH 3 ) has attracted interest as a future carbon-free synthetic fuel due to its economic storage and transportation. In this study, quasi direct numerical simulations (quasi-DNS) with detailed-chemistry have been performed in 3-D to examine the flame thickness and assess the validity of Damköhler’s first hypothesis for premixed turbulent planar ammonia/air and methane/air flames under different turbulence levels. The Karlovitz number is systematically changed from 4.26 to 12.06 indicating that all the test conditions are located within the thin reaction zones combustion regime. Results indicate that the ensemble average values of the preheat zone thickness deviate slightly from the thin laminar flamelet assumption, while the reaction zone regions remain relatively intact. Following the balance equation of reaction progress variable gradient, normal strain rate and the tangential diffusion component of flame displacement speed variation in the normal direction to the flame surface are found to be responsible for thickening the flame. However, the sum of reaction and normal diffusion components of flame displacement speed variation in the normal direction to the flame surface is in charge of flame thinning for ammonia/air and methane/air flames. In addition, the validity of Damköhler’s first hypothesis is confirmed by indicating that the ratio of the turbulent burning velocity to the unstrained premixed laminar burning velocity is relatively equal to the ratio of the wrinkled to the unwrinkled flame surface area. Furthermore, the probability density functions of the density-weighted flame displacement speed show that the bulk of flame elements propagate identical to the unstrained premixed laminar flame.

show abstract

Section: Reactive Flow Solvermentioning

confidence: 99%

Section: Reactive Flow Solvermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inner Flame Front Structures and Burning Velocities of Premixed Turbulent Planar Ammonia/Air and Methane/Air Flames

Tamadonfar

Karimkashi

Kaario

et al. 2022

Flow Turbulence Combust

Self Cite

View full text Add to dashboard Cite

show abstract

“…3 Nevertheless, high-fidelity reactive flow simulations can be still feasible through efficient usage of the provided computational resources, in addition to cell-based optimization of the chemistry solution. 4 In general, considerable efforts have been made in the literature to improve the computational performance of CFD simulations involving turbulence, [5][6][7][8][9] shock waves, 10 gas rarefaction, 11 geometrical parameterization, 12 or chemical reactions [13][14][15][16] through model reduction, machine learning, or efficient parallelization.…”

Section: Introductionmentioning

confidence: 99%

“…We have recently also introduced an open-source dynamic load balancing model called DLBFoam for OpenFOAM, providing speed-up up to a factor of 10. 4 In this work, we introduce a novel chemistry model that provides speed-up in reactive flow simulations in OpenFOAM by targeting the two major issues in reactive CFD simulations addressed above. We further extend our dynamic load balancing model DLBFoam, and focus on optimizing the cell-wise chemistry solution by introducing two new features.…”

Section: Introductionmentioning

confidence: 99%

Fast reactive flow simulations using analytical Jacobian and dynamic load balancing in OpenFOAM

et al. 2022

Self Cite

View full text Add to dashboard Cite

Detailed chemistry-based computational fluid dynamics (CFD) simulations are computationally expensive due to the solution of the underlying chemical kinetics system of ordinary differential equations (ODEs). Here, we introduce a novel open-source library aiming at speeding up such reactive flow simulations using OpenFOAM, an open-source software for CFD. First, our dynamic load balancing model by Tekg€ ul et al. ["DLBFoam: An open-source dynamic load balancing model for fast reacting flow simulations in OpenFOAM," Comput. Phys. Commun. 267, 108073 (2021)] is utilized to mitigate the computational imbalance due to chemistry solution in multiprocessor reactive flow simulations. Then, the individual (cell-based) chemistry solutions are optimized by implementing an analytical Jacobian formulation using the open-source library pyJac, and by increasing the efficiency of the ODE solvers by utilizing the standard linear algebra package. We demonstrate the speed-up capabilities of this new library on various combustion problems. These test problems include a two-dimensional (2D) turbulent reacting shear layer and three-dimensional (3D) stratified combustion to highlight the favorable scaling aspects of the library on ignition and flame front initiation setups for dual-fuel combustion. Furthermore, two fundamental 3D demonstrations are provided on nonpremixed and partially premixed flames, viz., the Engine Combustion Network Spray A and the Sandia flame D experimental configurations, which were previously considered unfeasible using OpenFOAM. The novel model offers up to two orders of magnitude speed-up for most of the investigated cases. The openly shared code along with the test case setups represent a radically new enabler for reactive flow simulations in the OpenFOAM framework.

show abstract

A Numerical Study on Premixed Turbulent Planar Ammonia/Air and Ammonia/Hydrogen/Air Flames: An Analysis on Flame Displacement Speed and Burning Velocity

Tamadonfar

Karimkashi

Kaario

et al. 2023

Flow Turbulence Combust

View full text Add to dashboard Cite

The economic storage and transportation of ammonia ($$\hbox {NH}_3$$ NH 3 ), and its capability to be thermally decomposed to hydrogen ($$\hbox {H}_2$$ H 2 ) make it a potential carbon-free synthetic fuel for the future. To comprehend the fundamental characteristics of $$\hbox {NH}_3$$ NH 3 as a primary fuel enriched with $$\hbox {H}_2$$ H 2 under low turbulent premixed flame conditions, three quasi direct numerical simulations (quasi-DNS) with detailed chemistry and the mixture-averaged transport model are conducted under decaying turbulence herein. The Karlovitz number is fixed to 4.28 for all the test conditions. The blending ratio ($$\alpha$$ α ), specifying the hydrogen concentration in the ammonia/hydrogen mixture, varies from 0.0 to 0.6. The results reveal that the mean value of the density-weighted flame displacement speed ($$S_{\textrm{d}}^{*}$$ S d ∗ ) is similar to (higher than) the unstrained premixed laminar burning velocity ($$S_{\textrm{L}}^{0}$$ S L 0 ) for $$\hbox {NH}_3/$$ NH 3 / air flame ($$\hbox {NH}_3/\hbox {H}_2/$$ NH 3 / H 2 / air flames). Furthermore, the performance of two extrapolation relations for estimating $$S_{\textrm{d}}^{*}$$ S d ∗ as linear and non-linear functions of flame front curvature is discussed thoroughly. The performances of both models are almost similar when evaluating the data near the leading edge of the flame. However, the non-linear one offers more accurate results near the trailing edge of the flame. The results show that the mean flame stretch factor increases with increasing the blending ratio, suggesting that the mean flamelet consumption velocity deviates from $$S_{\textrm{L}}^{0}$$ S L 0 by enriching the mixture with $$\hbox {H}_2$$ H 2 . The mean value of the local equivalence ratio ($$\phi$$ ϕ ) for the turbulent $$\hbox {NH}_3/$$ NH 3 / air flame is almost equal to its laminar counterpart, while it deviates significantly for $$\hbox {NH}_3/\hbox {H}_2/$$ NH 3 / H 2 / air flames. In addition, the local equivalence ratio for the flame front with positive curvature values is higher than the negatively curved regions for $$\hbox {NH}_3/\hbox {H}_2/$$ NH 3 / H 2 / air flames due to $$\hbox {H}_2$$ H 2 preferential diffusion. Furthermore, the results indicate that hydrogen is consumed faster in positively curved regions compared to the negatively curved zones due to enhanced reaction rates of specific chemical reactions.

show abstract

DLBFoam: An open-source dynamic load balancing model for fast reacting flow simulations in OpenFOAM

Cited by 49 publications

References 24 publications

Inner Flame Front Structures and Burning Velocities of Premixed Turbulent Planar Ammonia/Air and Methane/Air Flames

Inner Flame Front Structures and Burning Velocities of Premixed Turbulent Planar Ammonia/Air and Methane/Air Flames

Fast reactive flow simulations using analytical Jacobian and dynamic load balancing in OpenFOAM

A Numerical Study on Premixed Turbulent Planar Ammonia/Air and Ammonia/Hydrogen/Air Flames: An Analysis on Flame Displacement Speed and Burning Velocity

Contact Info

Product

Resources

About