Advanced Devolatilisation model for Lagrangian particles in OpenFOAM

Loading...
Thumbnail Image
Date
Authors
Futschik, Christian
Keyword
CFD , coal , dual competing devolatilisation rate model , lagrangian particle
Abstract
To study the devolatilisation process associated with coal combustion, computational fluid dynamics (CFD) is a very useful tool. The aim of this thesis is to accurately model coal devolatilisation at high temperatures associated with cement kilns to facilitate the further study of coal combustion within a cement kiln, to increase combustion efficiency thereby decreasing the environmental impact associated with cement production. The devolatilisation process can be modeled using two main types of models: the global and the structural model. Global models use parameters found by a simple kinetic analysis whereas the structural models require thorough analysis of the coal particles chemical structure. Due to it's simplicity and ability to model all types of coal without relying on in-depth analytical methods and parameter predictions for unknown coal types, the global model of the 'dual-competing rate' type was used. This model was implemented and optimized in the open-source CFD software OpenFOAM to allow increased accuracy when examining high temperature, high heating rate, and/or multi-peak rate volatile release coal devolatilisation. It has proven to be effective in high temperature and heating rate scenarios (such as those found in a cement kiln) and when modelling multi-peak rate volatile releases. Additionally, the particle specific heat (cp) model used within the original OpenFOAM code was modified as it does not allow for particle phase or specie mass fraction modifications without altering the particle cp value. Due to the high working temperatures, the Kirov cp model is used. The implementation of both models have shown an increase in high temperature devolatilisation dynamics and heating rate accuracy in addition to an increase in modelling multi-peak rate volatile release accuracy by a factor of 5 when compared to the single kinetic rate model.
External DOI