Combining flamelet-generated manifold and machine learning models in simulation of a non-premixed diffusion flame

Li, Kaimeng; Rahnama, Pourya; Novella, Ricardo; Somers, Lmt Bart

doi:10.1016/j.egyai.2023.100266

Cited by 7 publications

(2 citation statements)

References 46 publications

(49 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is attributed to complex chemical mechanisms involving thousands of chemical species and reactions. Furthermore, a wide variety of time and length scales present in the simulation of the combustion process, particularly for turbulent flames [28], lead to a stiff system of equations that increase the computational complexity. Despite many challenges, consideration of detailed chemical kinetics is essential to obtain high-fidelity predictions of the combustion process for specific industrial applications, such as IC engines.…”

Section: Flamelet-generated Manifold Combustion Modelmentioning

confidence: 99%

Tabulated Chemistry Combustion Model for Cost-Effective Numerical Simulation of Dual-Fuel Combustion Process

Stipic,

Basara,

Schmidt

et al. 2023

Energies

View full text Add to dashboard Cite

This study is dedicated to improving the efficiency of the flamelet-generated manifold (FGM) tabulated chemistry combustion modeling approach for predicting the combustion process in diesel-ignited internal combustion (IC) engines. The primary focus is on reducing table generation time and memory requirements. To accurately predict dual-fuel combustion processes, it is important to model both premixed and non-premixed combustion regimes. However, attempting to include both regimes in a single FGM lookup table leads to significant increases in the table size and generation time. In response, this work proposes a dual-table configuration, with each table dedicated to a specific regime. The solution is then interpolated from these tables based on the calculated combustion regime indicator during the computational fluid dynamics (CFD) simulation. This approach optimizes computational efficiency while ensuring an accurate representation of dual-fuel combustion. Additionally, to establish a cost-effective and accurate 3D CFD simulation workflow, the dual-table FGM methodology is coupled with the partially averaged Navier–Stokes (PANS) turbulence model. The feasibility of the proposed FGM methodology is tested utilizing six chemical kinetics mechanisms with different levels of detail. The results of this study demonstrated that the dual-table approach significantly accelerates table generation time and reduces memory requirements compared to a single table that includes both combustion regimes. Furthermore, 3D CFD simulation results of the dual-fuel combustion process are validated against available experimental data for three engine operating points. The in-cylinder pressure traces and rate of heat release obtained from the 3D CFD simulations employing the FGM PANS methodology show good agreement with experimental measurements, confirming the accuracy and reliability of this modeling approach.

show abstract

Section: Flamelet-generated Manifold Combustion Modelmentioning

confidence: 99%

Tabulated Chemistry Combustion Model for Cost-Effective Numerical Simulation of Dual-Fuel Combustion Process

Stipic,

Basara,

Schmidt

et al. 2023

Energies

View full text Add to dashboard Cite

show abstract

“…However, research on this approach has so far been limited to simple cases, necessitating further extensions and investigations. Moreover, several studies also combine machine learning methods and traditional approaches to tackle the complex fuel problems, including using a data-driven method to reduce the full chemistry to a subspace manifold by linear and non-linear models − and using neural networks to predict the flamelet-generated manifolds. − …”

Section: Introductionmentioning

confidence: 99%

Clustering-Enhanced Deep Learning Method for Computation of Full Detailed Thermochemical States via Solver-Based Adaptive Sampling

Chen,

Mehl,

Faney

et al. 2023

Energy Fuels

View full text Add to dashboard Cite

Detailed chemistry computations are indispensable in numerous complex simulation tasks, which focus on accurately capturing the ignition process or predicting pollutant levels. The machine learning method is a modern data-driven approach for predicting a full detailed thermochemical state-to-state behavior in reacting flow simulations. By combining unsupervised clustering algorithms to subdivide the composition space, the complexity of adaptive regression models for temporal dynamics can be significantly reduced. In this article, a more compact dataset is generated, which is essential for the clustering algorithm, by leveraging the adaptive CVODE solver time steps for data augmentation for stiff reactive states. A learning workflow that utilizes a deep residual network model (ResNet) in conjunction with an adaptive clustering algorithm is proposed. This approach aims to replace the stiff ordinary differential equation direct integration solver traditionally used for computing thermochemical species’ state-to-state temporal evolution for detailed chemistry simulations. The learning models are adaptively trained using the K-means clustering algorithm in the non-linear transformation space for different subspaces of dynamic systems. Three test cases, H2 (9 species), C2H4 (32 species), and CH4 (53 species), are investigated, each exhibiting varying complexities. The study demonstrates that the iterative predictions of thermochemical states align well with the results obtained from direct numerical integration. Additionally, employing multiple adaptive regression models in subdomains yields superior performance compared to a single regression model prediction case.

show abstract

Modeling of water gas shift reaction using neural network trained on detailed kinetic mechanisms

Yamaguchi,

Matsukawa,

Numazawa

et al. 2024

Chemical Engineering Journal

View full text Add to dashboard Cite

Combining flamelet-generated manifold and machine learning models in simulation of a non-premixed diffusion flame

Cited by 7 publications

References 46 publications

Tabulated Chemistry Combustion Model for Cost-Effective Numerical Simulation of Dual-Fuel Combustion Process

Tabulated Chemistry Combustion Model for Cost-Effective Numerical Simulation of Dual-Fuel Combustion Process

Clustering-Enhanced Deep Learning Method for Computation of Full Detailed Thermochemical States via Solver-Based Adaptive Sampling

Modeling of water gas shift reaction using neural network trained on detailed kinetic mechanisms

Contact Info

Product

Resources

About