Experimental and numerical study of particle velocity distribution in the vertical pipe after a 90° elbow

Abstract

A major problem of utilizing co-firing technique is controllably distributing fuel mixtures of pulverized coal and granular biomass in a common pipeline. This research related into particle velocity distribution in the vertical pipe after a right angle elbow was undertaken using the Laser Doppler Anemometry (LDA) technique and a coupled computational fluid dynamics (CFD)-discrete element method (DEM) simulation. According to the similarity criterion of Stokes Number, three types of glass beads were used to model the dilute gas-solid flow of pulverized coal or biomass pellets. In the vertical pipe after an elbow (R/D=1.3, R is the bend radius as 100mm and D is pipe diameter as 75mm), a horseshoe shape feature has been found on cross-sectional distributions of the axial particle velocity for all three types of glass beads on the first section which is 15mm away from the blend exit. At the further downstream sections, the horseshoe-shaped feature is gradually distorted until it completely disappears. The distance for total disintegration is about 300mm away from the first section for the first type of glass beads, 150mm away for the second type and 75mm away for the third one. As for particle number rate, its cross-sectional distribution on the first section shows that a rope is formed at the pipe outer wall where the maximum particle number rate occurs. On the whole, the rope formed by the first type of glass beads can be still observed at the section which is 450mm away from the first section. For the second type of glass beads, the rope will disintegrate from the section (300mm away) according to the cross-sectional distributions of the dimensional value of particle axial velocity divided by air conveying velocity. Overall, the roping characteristic is not obviously shown in the gas-solid flow of the third type of glass beads after the first section. All of these indicate that ropes formed from larger particles disperse more easily, for reasons perhaps related to their higher inertia. In addition, CFD-DEM analysis was employed to determine the particle characteristics to confirm this assumption. The numerical results indicated that the flow pattern has the highest axial velocity near the elbow to form the ‘horseshoe’ pattern. As the particles with lower Stokes Number were easier to follow the air flow, this is the reason why the horseshoe pattern of smaller particles was more obvious than larger particles.