Numerical investigation of influence on heat transfer characteristics to pneumatically conveyed dense phase flow by selecting models and boundary conditions

Abstract

Solids moving with a gas stream in a pipeline can be found in many industrial processes, such as power generation, chemical, pharmaceutical, food and commodity transfer processes. A mass flow rate of the solids is important characteristic that is often required to be measured (and controlled) to achieve efficient utilization of energy and raw materials in pneumatic conveying systems. The methods of measuring the mass flow rate of solids in a pneumatic pipeline can be divided into direct and indirect (inferential) measurements. A thermal solids' mass flow-meter, in principle, should ideally provide a direct measurement of solids flow rate, regardless of inhomogeneities in solids' distribution and environmental impacts. One key issue in developing a thermal solids' mass flow-meter is to characterize the heat transfer between the hot pipe wall and the gas-solids dense phase flow. The Eulerian continuum modeling with gas-solid two phases is the most common method for pneumatic transport. To model a gas-solid dense phase flow passing through a heated region, the gas phase is described as a continuous phase and the particles as the second phase. This study aims to describe the heat transfer characteristics between the hot wall and the gas-solids dense phase flow in pneumatic pipelines by modeling a turbulence gas-solid plug passing through the heated region which involves several actual and crucial issues: selections of interphase exchange coefficient, near-wall region functions and different wall surface temperatures. A sensitivity analysis was discussed to identify the influence on the heat transfer characteristics by selecting different interphase exchange coefficient models and different boundary conditions. Simulation results suggest that sensitivity analysis in the choice of models is very significant. The simulation results appear to show that a combination of choosing the Syamlal-O'Brien interphase exchange coefficient model and the standard k-ε model along with the standard wall function model might be the best approach, by which, the simulation data seems to be closest to the experimental results.