A new LTNE analysis process for predicting the steady-state temperature distribution in a fixed bed of randomly packed mono-sized rough spheres based on pore-scale simulations

Abstract

A new analysis process for the LTNE (local thermal non-equilibrium) model to solve for steady-state, fully-developed convection in mono-sized-sphere packed bed is proposed to predict the temperature distribution inside a fixed bed. For an actual mono-sized-sphere packed bed, the voidage lies between those of the regularly arranged beds of FCC, BCC, and SC. Through an interpolation/extrapolation process based on the pore-scale computation results of the SC, BCC, and FCC packings, the heat transfer parameters of a randomly packed bed can be calculated. To reliably estimate the local heat transfer rate, the regional mass flow rates in a packed bed are accurately predicted first. Then, the regional solid-phase thermal conduction can be independently evaluated without interference from the regional fluid-solid convection heat transfer. Additionally, a finite-thickness wall is constructed into the computational model to significantly improve the accuracy of the predicted radial temperature profiles. With realistic temperature self-adjustment within the wall, there is no need to set unrealistic boundary conditions of either isothermal, isoflux or their combinations at the outer face of the porous zone, as many conventional methods in the literature do. The solution process and necessary parameters are summarized into flowcharts. Finally, the accuracy of the new LTNE analysis process in predicting the temperature distribution throughout the packed bed is validated with the literature experiment. In contrast, the conventional methods fail to provide the correct trend of the temperature variation.