Bulletin of the American Physical Society
76th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 19–21, 2023; Washington, DC
Session J40: Reacting Flows: Computational Methods |
Hide Abstracts |
Chair: Wayne Strasser, Liberty University Room: 204C |
Sunday, November 19, 2023 4:35PM - 4:48PM |
J40.00001: Accelerating Exascale Reacting Flow Simulations with a Symbolic Analytic Jacobian formulation for QSSA Mechanisms Nicholas T Wimer, Malik Hassanaly, Lucas Esclapez, Anne Felden, Julia A Ream, Marc T Henry de Frahan, Jon Rood, Marc Day One of the most time-consuming aspects of modern reacting flow simulations of real-world engineering application is the evaluation of the chemical reaction mechanism. In this talk we evaluate the recent implementation of a full symbolic analytic Jacobian method in the exascale CFD software Pele on a dual-fuel multi-pulse jet simulation. We compare the results and performance metrics using a range of options in the Pele suite of codes and evaluation of a skeletal and quasi-steady-state approximation chemical reaction mechanism. The methods presented here are deployable on a range of computational architectures, including GPUs, and are scalable up to exascale computing systems. |
Sunday, November 19, 2023 4:48PM - 5:01PM |
J40.00002: An implementation of heterogeneous chemistry into AMReX-based combustion solvers Anirudh Jonnalagadda, Shashank Tiwari, Konduri Aditya The ability to model heterogeneous combustion processes using modern exascale HPC compute architectures holds the potential to address demanding challenges in the energy sector. In this work, the exascale computing project's Pele suite of codes for reacting flows (https://amrex-combustion.github.io/) has been extended to include heterogeneous combustion modelling capabilities. The required heterogeneous chemical kinetics calculations are performed through subroutines that are automatically generated via a reworked implementation of the CEPTR module (https://amrex-combustion.github.io/PelePhysics/Ceptr.html) within the PelePhysics library (https://amrex-combustion.github.io/PelePhysics/). Specifically, CEPTR is equipped to handle three kinds of interface reactions (elementary, surface-coverage-modified, and arbitrary forward reaction rates) along with sticking reactions with and without the inclusion of the Motz-Wise correction. Further, a new surface module has been included in PelePhysics to interface the heterogeneous subroutine library with application codes. In order to verify, validate and benchmark the new additions, a new finite-gap stagnation flow solver with heterogeneous reactions has been developed within the AMReX-combustion framework. The solver numerically solves the steady state stagnation flow similarity equations using the SUNDIALS ARKODE library (https://computing.llnl.gov/projects/sundials/arkode) and is validated through quantitative comparisons of the axial, radial velocity and temperature profiles, along with key species mass fraction profiles along the axial direction against Cantera's impinging jet solver for the case of H2 assisted catalytic combustion of CH4 on Pt. The current work paves the way for enabling 3D heterogeneous combustion simulations within the AMReX-combustion framework. |
Sunday, November 19, 2023 5:01PM - 5:14PM |
J40.00003: Hybrid parallelization of In-Situ Adaptive Manifolds for computationally efficient turbulent combustion simulations Stephen T Fush, Israel J Bonilla, Michael E Mueller The cost of turbulent combustion simulations can be decreased by utilizing manifold-based combustion models, which project the thermochemical state onto a lower-dimensional manifold such that the thermochemical state can be computed separately from the flow solver. Traditionally, the manifold equation solutions for the thermochemical state were precomputed and pretabulated, resulting in large memory requirements and significant precomputation cost. In-Situ Adaptive Manifolds (ISAM) allows for solutions to the manifold equations to be computed as the flow simulation progresses and stored using In-Situ Adaptive Tabulation (ISAT), allowing for more general models to be used. The initial population of the ISAT databases can be accelerated by trading MPI processes for OpenMP threads due to faster computing of the solutions to the manifold equations and reduced redundancy in computing these solutions. However, once the ISAT databases are fully populated, the serialized ISAT algorithm leads to an overall increase in the computational cost of ISAM. In this work, a new adaptively threaded version of the ISAT algorithm has been developed to effectively leverage the OpenMP threads. Overall efficiency and scaling with the number of OpenMP threads will be demonstrated. |
Sunday, November 19, 2023 5:14PM - 5:27PM |
J40.00004: Influence of real gas effects on chemical kinetics in oxycombustion in supercritical carbon dioxide Mohammad J Rahimi, Marc T Henry de Frahan, Olga Doronina, Shashank Yellapantula, Ian Cormier, Marc Day, Michael J Martin, Bruce A Perry Oxycombustion in supercritical carbon dioxide is an integral part of the Allam Cycle, a technology that enables carbon-neutral use of fossil-fuels and carbon-negative use of biofuels. We simulate oxycombustion in a realistic combustor geometry for two sets of fuel conditions: pure methane, and a 40 percent methane/60 percent carbon dioxide blend, both at 343.15 K. The fuel jet mixes with a preheated swirler of 20 percent oxygen and 80 percent carbon dioxide at 1005.35 K, with a 100 percent carbon dioxide coflow at 783.15 K. The entire system operates at a pressure of 300 bar, putting the entire system above the critical temperature and pressure of carbon dioxide. Simulations are performed using PeleC, a compressible block-structured adaptive-mesh refinement (AMR) reacting flow code. Direct comparison of results using thermodynamically self-consistent implementations of the ideal gas equation of state and the Soave-Redlich-Kwong (SRK) equation of state allow quantification of the real gas impacts on combustion kinetics, flow temperatures, and pollutant formation. |
Sunday, November 19, 2023 5:27PM - 5:40PM |
J40.00005: Adaptive time step control using reinforcement learning for stiff chemical kinetic systems Vijayamanikandan Vijayarangan, Harshavardhana A Uranakara, Francisco E Hernandez Perez, Hong G Im This work focuses on identifying an adaptive time step size for the integration of stiff chemistry ordinary differential equations (ODEs) encountered in reactive flow simulations. Identifying the right time step size in high-fidelity numerical simulations is an expensive task, which requires eigen-decomposition of the chemical Jacobian to identify the smallest timescales. An incorrect specification of time step size results in numerical instability and eventually leads to a blowup of the ODE integrator. To tackle these challenges, backward difference ODE integrators have been developed, which rely on a trial-and-error or predetermined criterion-based time-step control process. This study employs machine learning techniques for time step size control in explicit integration methods for stiff chemical systems. The proposed approach involves a model-based Reinforcement Learning (RL) algorithm to learn an optimized policy for time step control in an explicit ODE integrator. The RL agent's primary goal is to minimize interactions with the environment while maintaining near-optimal performance through efficient time step size control. The effectiveness of this method is evaluated through tests involving a constant pressure batch reactor with H2-air mixture and contrasted with time step size control algorithms used for integrating the system with stiff chemical kinetics. The results highlight the potential of the RL-based time step size control technique to enhance the performance of explicit integration methods in the realm of chemical kinetics and reacting flow simulations. |
Sunday, November 19, 2023 5:40PM - 5:53PM |
J40.00006: Matrix Product State Simulation of Reactive Flows Robert Pinkston, Peyman Givi, Juan José Mendoza Arenas, Nikita Gourianov, Dieter Jaksch The matrix product state (MPS) representation, developed for approximating quantum many-body systems, exploits their correlation structure to accurately capture the underlying physics in a low-rank form (i.e., in a massively reduced state space). Here, the methodology is invoked to simulate several reactive flow systems. The flow fields and governing differential operators are recast in the context of MPS, and their dynamics is simulated for various degrees of truncation. The generated MPS-reduced order solutions are appraised against the results obtained via direct numerical simulation of the original equations. The computational complexity of MPS is assessed to determine its potential for calculating complex multi-scale turbulent reacting flows. |
Sunday, November 19, 2023 5:53PM - 6:06PM |
J40.00007: Numerical methodology for simulating premixed flame propagation within closed vessels Gautham Krishnan, Hang Yu, Carlos Pantano, Moshe Matalon A Navier-Stokes/embedded-manifold numerical methodology to simulate the propagation of premixed flames in closed vessels will be discussed. The methodology is based on the hydrodynamic theory wherein the flame is confined to a surface that propagates at a speed that has been derived by considering the transport and chemical processes occurring inside the flame zone. Unlike freely propagating flames, which evolve under nearly isobaric conditions, the mean pressure in the vessel increases in time, the gas is compressed, its temperature rises, and the burning rate increases substantially. The numerical solver we developed can handle these temporal variations in laminar and turbulent flows, and using an immersed boundary method, it accurately assures the implementation of the boundary conditions for vessels of arbitrary geometry. Results of flame propagation in rectangular and spherical (circular) vessels will be presented and compared to analytical solutions. Future studies will address multi-dimensional flames, where fluctuations of the flame surface result from instabilities and/or turbulence. |
Sunday, November 19, 2023 6:06PM - 6:19PM |
J40.00008: Implementation of Complex Free Radical Polymerization Chemistry into a Plant-Scale Computational Fluid Dynamics Reactor Model Elijah P Yoder, Wayne Strasser, Victor Lin Simulation of chemical reactors has become popular for predicting the resulting polymer product properties. Many detailed polymer property simulation packages exist; however, they assume uniform mixing in the reactor, omitting many physical phenomena. Computational Fluid Dynamics (CFD) provides more insight into the nonhomogeneity of the polymer reaction path, as it can simulate non-ideal mixing in reactors. An autoclave Low-Density Polyethylene (LDPE) reactor was used to verify the implementation of the complex polymerization chemistry. The first three steps utilize a Continuous Stirred Tank Reactor (CSTR) that assumes complete mixing, where chemical complexity was progressively increased. The results were then compared to an industry-standard polymer software package that also assumes a homogeneous mixture. In the fourth step, the complex chemistry was incorporated into a plant-scale reactor model and the results were compared to the plant reactor. For the first three steps, the CFD model only differed from the industry-standard software by 0.5%, 2.15%, and 1.75%, respectively. In the fourth step, the CFD model differed from the plant reactor by only 5%. This verified CFD model will allow future studies on the influence of inlet conditions, stirrer geometry, variable material properties, and others on the end polymer properties. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700