Online parameter inference for the simulation of a Bunsen flame using heteroscedastic Bayesian neural network ensembles

TL;DR

This paper introduces a heteroscedastic Bayesian neural network ensemble method for real-time inference of G-equation model parameters in Bunsen flames, enabling fast, online, and uncertainty-aware combustion simulations.

Contribution

It presents a novel application of heteroscedastic Bayesian neural networks for online parameter inference in flame modeling, reducing computational costs compared to traditional methods.

Findings

Accurately infers flame parameters in real-time

Provides uncertainty estimates comparable to ensemble Kalman filter

Achieves faster computation for combustion simulation

Abstract

This paper proposes a Bayesian data-driven machine learning method for the online inference of the parameters of a G-equation model of a ducted, premixed flame. Heteroscedastic Bayesian neural network ensembles are trained on a library of 1.7 million flame fronts simulated in LSGEN2D, a G-equation solver, to learn the Bayesian posterior distribution of the model parameters given observations. The ensembles are then used to infer the parameters of Bunsen flame experiments so that the dynamics of these can be simulated in LSGEN2D. This allows the surface area variation of the flame edge, a proxy for the heat release rate, to be calculated. The proposed method provides cheap and online parameter and uncertainty estimates matching results obtained with the ensemble Kalman filter, at less computational cost. This enables fast and reliable simulation of the combustion process.