Pulverised biomass and coal co-firing simulation using computational fluid dynamics :A numerical investigation into the aerodynamics of non-spherical particles and full scale combustion for pulverised fuel applications

Recent national and international emissions legislation, in particular sulphur-dioxide, and the rapid depletion of fossil fuels are forcing power producing industries to look at various alternatives, such as biomass and co-firing techniques. Biomass may be transported to the burners of a pulverised fuel (PF) boiler either mixed with the primary fuel, in general coal, or used in dedicated pipelines. In both cases, the transportation of biomass is different due to its composition, size and shape to the transportation of coal.

This thesis investigates the computational modelling techniques for a biomass and biomass blend particle transportation (arboreal and flour) in a pipeline with a transverse elbow, the three-phase flow of a coal and biomass co-fire blend in the primary air annulus of a swirl burner and the combustion of a coal and pelletised straw mixture in a full scale furnace using dedicated burners for the biomass injection.

The comparison of spherical and non-spherical drag models, under gravity, as well as Saffman lift, inter-particle collision and randomised impulsive wall collision models has been investigated. Good agreement was observed between the computational fluid dynamics (CFD) simulations and the experimental data, using a non-spherical drag model. In both cases, due to the dilute volume fraction and secondary air flow, inter-particle collisions and lift were insignificant. In the annulus, lateral regions of high particle concentration were predicted, which are not observed physically.

Numerical simulations of a 300MWe tangentially fired furnace, co-firing bituminous coal and pelletised straw, have been performed and compared to experimental data. Bituminous coal was co-fired with pelletised straw. Good agreement was obtained between the CFD predictions and the experimental data so that the trends of furnace temperature, NOx emissions and carbon burnout reduction, as biomass load is increased, were observed. Quantitative prediction of unburnt carbon (UBC) and NOx require a more detailed picture of the processes within the furnace at higher temperatures than that currently provided by experimental data.