Calculation of premixed combustion within inert porous media with model parametric uncertainty quantification

Abstract

The present study focuses on uncertainties existing in porous media parameters and in the inlet reactant mixture conditions of solid oxide fuel cell off-gas combustion. Propagation of uncertainty from the model input parameters to the output stochastic variables is quantified using a non-intrusive spectral projection method based on polynomial chaos expansion. The non-intrusive nature of this method allows the solution of the stochastic problem to be obtained directly from the deterministic model without requiring modification of the governing equations. Quantification of uncertainty is investigated in a one-dimensional model for premixed combustion within inert porous media. The model includes detailed chemistry and solves the gas- and solid-phase energy balances coupled by convective heat exchange, including radiative heat transfer in the solid-phase. The results denote that the uncertainties in the porous media heat transfer parameters are relevant and originate a relatively high error bar on the CO emission and burning velocity. When the inlet reactant mixture uncertain conditions is also accounted for, it overcomes the influence of the other uncertain parameters on the gas- and solid-phase temperatures error bar. Both types of parametric uncertainty sources (inlet conditions and porous media parameters) are important in order to establish the error bar on the CO emission and burning velocity predictions.