Acceleration of Chemical Kinetics Computation with the Learned Intelligent Tabulation (LIT) Method

Abstract

In this work, a data-driven methodology for modeling combustion kinetics, Learned Intelligent Tabulation (LIT), is presented. LIT aims to accelerate the tabulation of combustion mechanisms via machine learning algorithms such as Deep Neural Networks (DNNs). The high-dimensional composition space is sampled from high-fidelity simulations covering a wide range of initial conditions to train these DNNs. The input data are clustered into subspaces, while each subspace is trained with a DNN regression model targeted to a particular part of the high-dimensional composition space. This localized approach has proven to be more tractable than having a global ANN regression model, which fails to generalize across various composition spaces. The clustering is performed using an unsupervised method, Self-Organizing Map (SOM), which automatically subdivides the space. A dense network comprised of fully connected layers is considered for the regression model, while the network hyper parameters are optimized using Bayesian optimization. A nonlinear transformation of the parameters is used to improve sensitivity to minor species and enhance the prediction of ignition delay. The LIT method is employed to model the chemistry kinetics of zero-dimensional H2–O2 and CH4-air combustion. The data-driven method achieves good agreement with the benchmark method while being cheaper in terms of computational cost. LIT is naturally extensible to different combustion models such as flamelet and PDF transport models.

A Data-Based Hybrid Chemistry Acceleration Framework for the Low-Temperature Oxidation of Complex Fuels

“…And while a very deep and wide network can be used to approximate the entire dynamics, this may lead to higher computational costs. As a result, it is necessary to cluster data to improve accuracy and lower computational costs [17,18,31,32]. Self-organizing maps (SOMs) [33] and K-means [34] are well-established clustering techniques for clustering chemical data [17][18][19].…”

“…As a result, it is necessary to cluster data to improve accuracy and lower computational costs [17,18,31,32]. Self-organizing maps (SOMs) [33] and K-means [34] are well-established clustering techniques for clustering chemical data [17][18][19]. Local principal component analysis (LPA) [35] is another approach used to cluster data with similar chemical compositions.…”

“…These representative species can then be modeled using ANNs. ANNs have been successfully used in combustion for the tabulation of chemical reactions and chemistry reduction [17][18][19][20][21]. Additionally, as a regression tool for species and temperature chemical source terms, ANNs have been used to accelerate the evaluation of chemistry [22][23][24][25][26][27].…”

Efficient neural network models of chemical kinetics using a latent asinh rate transformation

“…1,2,8,9 This can be done by tabulating the kinetics or even a time integration step. 9,10 Latter is often done for gas-phase reactive systems [11][12][13] because the integration of the stiff ODE system resulting from the gasphase kinetics is very time-consuming. For heterogeneous catalysis, timescales of surface reactions and the gas phase are usually separable via the steady state approximation.…”

Application of dense neural networks for manifold-based modeling of flame-wall interactions

“…Again, the ANN showed a comparable accuracy to a tabulated manifold with acceptable computational overhead for LES applications. Recently, several investigations have been carried out applying machine learning in reactive flow simulations for manifold representation [7][8][9][10][11][12][13], turbulence-chemistry interactions [14][15][16][17] and modeling chemical kinetics [18][19][20][21][22]. This was made possible by the development of open-source deep learning frame-works that followed the breakthrough of deep learning in the field of computer vision in the past decade.…”