Abstract
Flow thermomechanics in reactive porous media is of importance in industry including the thermal processing of fossil fuel (coking understood as a slow pyrolysis) involving devolatilisation. On the way to provide a detailed description of the process, a multi-scale approach was chosen to estimate effective transport coefficients. For this case the Lattice Boltzmann method (LBM) was used due to its advantages to accurately model multi-physics and chemistry in a random geometry of granular media. After account for earlier studies, the paper presents description of the model with improved boundary conditions and a benchmark case. Results from meso-scale LBM calculations are presented and discussed regarding the spatial resolution and the choice of relaxation parameter along its influence on the accuracy compared with empirical formulae. Regarding the estimation of effective thermal conductivity coefficient it is shown that occurrence of devolatilization has a crucial effect by reducing heat transfer. Some quantitative results characterise the propagation of thermal front; also presented is the evolution of effective thermal conductivity. The work is a step forward towards a physically sound simulation of thermal processing of fossil fuel.
Highlights
One of the challenges in numerical modelling of technological processes in granular media is an accurate and efficient handling of random geometry
In (Asinari et al 2007), the Authors analyse phenomena observed in a solid oxide fuel cell (SOFC) at the level of a single pore in a representative element of volume (REV) being a randomly created porous medium whose topology is closely related to that of SOFC
Heat transport in complex geometry was analyzed by Wang and Pan (2008), focused on the REV domain; discussion is mainly concentrated on the comparison of computational results with theoretical predictions
Summary
One of the challenges in numerical modelling of technological processes in granular media is an accurate and efficient handling of random geometry. In the cases where process conditions prevent from making detailed measurements, the computational modelling has become an important player Such investigations can be aided by so-called structure models proposed for estimation of effective thermal conductivity (or other transport coefficients) for geometries where the structure is ordered (like packed beds, successively arranged layers, etc.). The authors compared results from the LBM with implementations of finite volume method (FVM) in a commercial software as well as an in-house code In both works, the comparison of LBM results with other estimations (different numerical schemes, references and empirical correlations) substantiate the usefulness of LBM in modelling of heat transfer in complex geometry. One additional test case is presented for non-reactive flow in the same computational domain
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
