Gas-solid fluidized beds operating in the bubbling regime are widely used in chemical engineering and in the power industry. The huge number of particles involved in these systems makes the adoption of Lagrangian models not suitable for the simulation of devices of practical interest. As a consequence, the Eulerian-Eulerian two-phase approach is preferred. An implementation of this approach is offered in the application twoPhaseEulerFoam, presents in the OpenFOAM library. This implementation represents a base for several extensions proposed in the literature, like the BIOTC code or the biomassGasificationFoam code. However any extensive validation of the solver has been conducted up to now. Aim of this work is to verify the ability of the solver to correctly reproduce the experimental set-up proposed in Kuipers et al. (Powder Tech. 1991). Firstly, numerical parameters are investigated, to determine the grid spacing and time step requirements. Then the effect of the available models and closure correlations on the correct reproduction of the transient phenomenon will be illustrated. Finally several numerical caveats, as a proper selection of the linear solvers for the different equations and the scalability for parallel execution will be highlighted, giving a guidance for a successful integration.
Validation of the twoPhaseEulerFoam solver for jet inlet fluidized beds
F S Marra
2016
Abstract
Gas-solid fluidized beds operating in the bubbling regime are widely used in chemical engineering and in the power industry. The huge number of particles involved in these systems makes the adoption of Lagrangian models not suitable for the simulation of devices of practical interest. As a consequence, the Eulerian-Eulerian two-phase approach is preferred. An implementation of this approach is offered in the application twoPhaseEulerFoam, presents in the OpenFOAM library. This implementation represents a base for several extensions proposed in the literature, like the BIOTC code or the biomassGasificationFoam code. However any extensive validation of the solver has been conducted up to now. Aim of this work is to verify the ability of the solver to correctly reproduce the experimental set-up proposed in Kuipers et al. (Powder Tech. 1991). Firstly, numerical parameters are investigated, to determine the grid spacing and time step requirements. Then the effect of the available models and closure correlations on the correct reproduction of the transient phenomenon will be illustrated. Finally several numerical caveats, as a proper selection of the linear solvers for the different equations and the scalability for parallel execution will be highlighted, giving a guidance for a successful integration.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
