RT Journal Article ID 19beaa591e3dd16c A1 Epple, Bernd A1 Fiveland, Woody A1 Krohmer, Bernd A1 Richards, Galen A1 Benim, Ali Cemal T1 ASSESSMENT OF TWO-PHASE FLOW MODELS FOR THE SIMULATION OF PULVERIZED COAL COMBUSTION JF International Journal of Energy for a Clean Environment JO IJECE YR 2005 FD 2005-10-18 VO 6 IS 3 SP 267 OP 287 AB CFD procedures for dispersed two-phase flow modelling are mostly based on the Euler-Lagrange approach, which employs an Eulerian gas-phase and a Lagrangian particle-phase description. This modelling approach is also used in the field of pulverized coal combustion. An alternative modelling approach for the dispersed two-phase flow is the Euler-Euler approach, with the assumption of interpenetrating continua of gas and solid phases, i.e., the particulate phase is also modelled within an Eulerian frame of reference. In the present procedure called, EUPAC (EULERIAN PROCEDURE FOR PULVERIZED COAL COMBUSTION), which was implemented into a commercial CFD code, we adopt this type of modelling, assuming a fluid dynamical and a thermodynamical equilibrium between the phases. Within this framework, the treatment of the particulate phase becomes analogous to that of a species with a very large molecular weight in a single-phase mixture, resulting in a single set of transport equations for the "mixture." This approach is commonly known as a "heavy gas" approximation and is less expensive than the standard Euler-Lagrange approach, because the calculation of an inter-phase coupling is omitted. The reduction in the computer time is additionally enhanced, because the equation system becomes more robust and shows better convergence properties because of the missing problem of inter-phase coupling. For the prediction of the coal combustion, it is important to account for the impact of different particle sizes.
For the validation, the first step covers a calculation of a test flame in a laboratory-scale single-burner test facility. Starting with a no-swirl arrangement on a 2D-axisymmetric grid, the validation case was subsequently extended to a swirl flame in a 3D domain. Within this first validation step, special emphasis was put on the transfer of commonly available coal data (immediate and proximate analysis) to coal input parameters for the reaction modelling in the present approach.
In the second step, a full T-fired furnace of a boiler test facility with a plan area for the furnace of the first pass of 3.35 m by 2.9 m and 12.8 m high was used to assess the performance of the present modelling approach for industrial applications. The Euler-Euler results were compared to predictions carried out with an Euler-Lagrange formulation and a set of pilot scale measurements. PB Begell House LK https://www.dl.begellhouse.com/journals/6d18a859536a7b02,0a74cb660a6afac8,19beaa591e3dd16c.html