An integrated model for flexible simulation of biomass combustion in a travelling grate-fired boiler

Abstract

Grate-fired boiler is of considerable interest for burning biomass owing to its reduced sensitivity to variations in fuel composition and size. However, grate-fired technology exhibits unsatisfactory performance in terms of combustion efficiency, contaminant emissions, and flame stability. CFD modelling can provide reasonable optimizations to overcome these drawbacks and ultimately achieve clean and efficient energy production. Conventional approach simulates the fuel-bed and freeboard combustion separately using an in- and over-bed coupling procedure, where data of the gases leaving from the packed-bed top and the radiative heat flux emitted by the high-temperature flame onto the bed cannot be interacted in real-time. This may cause distinct deviations in the calculations. In this paper, a three-dimensional (3D) full-scale integrated model is developed for the simulation of grate-fired boiler, in which the intensive combustion of both the freeboard and fuel-bed is solved in one scheme using the Eulerian-Lagrangian method. The gas phase is considered as a continuous medium and calculated by the Eulerian model, while the solid particles are considered as a discrete phase and tracked via the Lagrangian method. The reliability of the model is well verified through various field measurements and observations. Results shows that the key parameters are non-uniformly distributed along the grate width, i.e., the aerodynamic and combustion environment is poor at the vicinity of water-cooled walls compared with that at the furnace center. This results in a significant lag in volatile gases release and char oxidation near the water-cooled wall on both sides, which finally causes obvious differences in the distribution of temperature and species. For instance, the peak of CO concentration at the center is 12.6 % at 3.3m from the feed entrance, while that near the water-cooled wall is 8.6 % around 3.5m. The char burnout ratio in the bottom ash can be accurately calculated by this model, with a negligible relative error between the simulation of 81.97 % and the measurement of 82.50 %. It also provides a novel method for the calculation of fly ash particles by using size-grouped and non-spherical particles. Small-size particles in the vicinity of the feed inlet, as well as a part of the particles burning at the end of the grate are entrained into the freeboard by the primary air supplied beneath the grate and the high-momentum secondary air on the bed top. This model can provide valuable theoretical guidance for optimizing the refined distribution of fuel and air in grate-fired power plants and mitigating irregular deposition (corrosion) on heat exchangers and water-cooled walls caused by fly ash.