Modeling the Vibrational Relaxation Rate Using Machine-Learning Methods

Bushmakova, M. A.; Кустова, Е. В.

doi:10.1134/s1063454122010022

Cited by 7 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of machine learning methods for modeling of physical systems has grown sharply, see [Fradkov, 2022;Plotnikov et al, 2019;Fradkov and Shepeljavyi, 2022]. Machine learning methods help to accurately predict physical quantities by processing large amounts of available data, which significantly reduces computational effort and allows for implementing detailed models of physical-chemical kinetics and transport processes [Istomin and Kustova, 2021;Campoli et al, 2022;Bushmakova and Kustova, 2022]. In our preliminary studies, we used neural networks to speed up the evaluation of specific heats, thermal conductivity, and shear viscosity for single-component gases and simple mixtures, which resulted in a speed-up ratio of up to 10 3 [Istomin and Kustova, 2021].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of state-specific transport properties using machine learning methods

Истомин

Кустова

Lagutin

et al. 2023

CAP

View full text Add to dashboard Cite

In this study, machine learning algorithms are employed to calculate state-to-state transport coefficients in nonequilibrium reacting gas flows. The focus is on the evaluation of thermal conductivity, shear viscosity, and bulk viscosity coefficients under conditions of strong coupling between vibrational-chemical kinetics and gas dynamics. In order to solve a regression problem for evaluating state-to-state transport coefficients, a specific software application with user interface is developed, which allows loading, processing, and saving of data arrays; configuring model architecture; training and evaluating models with various optimizers, loss functions, and metrics; making predictions using trained models. Using the developed software the multi-layer perceptron regression model is constructed and trained. The model is assessed in a binary mixture of molecular and atomic nitrogen taking into account 48 vibrational states; the coefficients are computed in the wide temperature range for the varying mixture composition. Good agreement of the results with the original transport coefficients calculated using rigorous but computationally expensive kinetic theory algorithms is shown. Applying machine learning techniques yields a significant speedup of about two orders of magnitude in the computation of transport coefficients. It is concluded that implementation of machine learning methods may considerably reduce the computational efforts required for nonequilibrium flow simulations.

show abstract

Section: Introductionmentioning

confidence: 99%

Evaluation of state-specific transport properties using machine learning methods

Истомин

Кустова

Lagutin

et al. 2023

CAP

View full text Add to dashboard Cite

show abstract

Investigation of non-equilibrium phenomena in nitrogen RF inductively coupled plasma discharges: a state-to-state approach

Kumar,

Munafò,

et al. 2024

J. Phys. D: Appl. Phys.

View full text Add to dashboard Cite

This work presents a vibrational and electronic (vibronic) state-to-state (StS) model for nitrogen plasmas implemented within a multi-physics modular computational framework to study non-equilibrium effects in inductively coupled plasma (ICP) discharges. The vibronic master equations are solved in a tightly coupled fashion with the flow governing equations eliminating the need for invoking any simplifying assumptions when computing the state of the plasma, leading to a high-fidelity physical modeling. The model’s computational complexity is reduced via a maximum entropy coarse-graining approach, verified through zero-dimensional isochoric calculations. The coarse-grained StS model is employed to study the plasma discharge in the ICP facility at the von Karman Institute (VKI) for Fluid Dynamics, Belgium. Results reveal pronounced discrepancies between StS predictions and those obtained based on local thermodynamic equilibrium (LTE) models, which are conventionally used in the simulation of such facilities. The analysis demonstrates a substantial departure of the internal state populations of atoms and molecules from the Boltzmann distribution. This has significant implications for energy coupling dynamics, affecting the discharge morphology. Further analysis reveals a quasi-steady-state population distribution in the plasma core, allowing for the construction of an efficient and “self-consistent” macroscopic two-temperature (2T) formulation. Non-LTE simulations indicate significant disparities between the StS model and the commonly used Park 2T model, whereas the newly proposed 2T model aligns closely with StS simulations, capturing key features of non-equilibrium plasma formation. In particular, the current study highlights the importance of the vibrational-translational (VT) energy transfer term in shapingthe plasma core morphology, suggesting a notable sensitivity to heavy-impact vibrational excitations and dissociative processes.

show abstract

Neural-Network-Based Approach to the Description of Vibrational Kinetics of Carbon Dioxide

Gorikhovskii

Кустова²

2022

Vestnik St.Petersb. Univ.Math.

View full text Add to dashboard Cite

Modeling the Vibrational Relaxation Rate Using Machine-Learning Methods

Cited by 7 publications

References 8 publications

Evaluation of state-specific transport properties using machine learning methods

Evaluation of state-specific transport properties using machine learning methods

Investigation of non-equilibrium phenomena in nitrogen RF inductively coupled plasma discharges: a state-to-state approach

Neural-Network-Based Approach to the Description of Vibrational Kinetics of Carbon Dioxide

Contact Info

Product

Resources

About