Numerical simulation of group combustion of pulverized coal

Abstract

A mathematical model for the group combustion of pulverized coal particles was developed in a previous work. It includes the Lagrangian description of the dehumidification, devolatilization and char gasification reactions of the coal particles in the homogenized gaseous environment resulting from the three fuels, CO, H 2 and volatiles, supplied by the gasification of the particles and their simultaneous group combustion by the gas phase oxidation reactions, which are considered to be very fast. This model is complemented here with an analysis of the particle dynamics, determined principally by the effects of aerodynamic drag and gravity, and its dispersion based on a stochastic model. It is also extended to include two other simpler models for the gasification of the particles: the first one for particles small enough to extinguish the surrounding diffusion flames, and a second one for particles with small ash content when the porous shell of ashes remaining after gasification of the char, non structurally stable, is disrupted.As an example of the applicability of the models, they are used in the numerical simulation of an experiment of a non-swirling pulverized coal jet with a nearly stagnant air at ambient temperature, with an initial region of interaction with a small annular methane flame. Computational algorithms for solving the different stages undergone by a coal particle during its combustion are proposed. For the partial differential equations modeling the gas phase, a second order finite element method combined with a semiLagrangian characteristics method are used. The results obtained with the three versions of the model are compared among them and show how the first of the simpler models fits better the experimental results.

Large Eddy Simulation of piloted pulverized coal combustion using the velocity-scalar joint filtered density function model

“…The laboratory scale pulverized coal jet flame studied at the Japanese Central Research Institute of Electric Power Industry (CRIEPI) [23,24] has already been simulated by many researchers [7,8,25,26] relatively detailed data of the process and is also used in this work. In the present work, the numerical results using the LES/VSJFDF, non-VSJFDF models, together with relatively simple sub-models for PCC were compared with the experimental data and Franchetti's numerical results [7].…”

Numerical study of HCl and SO2 impact on sodium emissions in pulverized-coal flames

“…As the sharp increase of computing capacity continues, computational fluid dynamics (CFD) 83 methods for the carrier-gas flow of PCC have evolved from Reynolds-averaged Navier-Stokes 84 (RANS) simulation (e.g. [24][25][26]) towards high-fidelity approaches, i.e., large-eddy simulation (LES, 85 e.g. [27][28][29][30][31][32][33][34]) and direct numerical simulation (DNS, e.g., [35][36][37]).…”

Identification of the initial particle size distribution for coal combustion simulations

“…Backreedy et al used the Rosin‐Rammler function to describe the PSD of pulverized coal and the size range of 25 × 10 −6 –200 × 10 −6 m was divided into 10 bins . Many researchers adopted the same method as Backreedy et al to set the initial PSD . Pedel et al suggested that two particles would behave significantly differently if their mass was not of the same order of magnitude, so several particle sizes across a large range should be chosen to represent the wide PSD .…”