MODELING OF DUST EXPLOSION IN LINKED PROCESS VESSELS

Abstract

Dust explosions have been a threat to mankind and property for centuries. The ignition and burning of solid particles in air is a function of many parameters such as the dispersion of the particles in the cloud, the particle size distribution and turbulence characteristics in the cloud. Dust explosion an interesting topic that has gained much interest in recent years is the propagation of an explosion between process units through pipes and ducts. An explosion in one of a series of connecting process units could spread to others by the ducts. The flame propagating in the duct tends to accelerate because of turbulence and results in a jet flame entering the second vessel. Due to this, high combustion rates are obtained at high pressures, even if the second vessel is vented and the amount of dust it contains does not present much danger in itself. Thus experimental investigation into explosions in connected vessels is not an easy task. In the present work Computational Fluid Dynamics (CID) code is used to analyze the phenomenon in the secondary vessel which is connected to primary vessel(in which the explosion has already occur) through a duct having height 0.5 m and length 2 m. The size of the particles and the height of the duct are variable. In the secondary vessel an amount of dust is initially dispersed uniformly at a volume fraction of 0.2. The geometry is shown in figure 3.1. It is shown by a result that the point in time at which an explosion can occur in the secondary vessel is a function of the duct height: for bigger duct height an explosion is more likely to occur, and will occur earlier. It is also a function of particle size, since the particles react differently to the pressure wave arising from the primary explosion and also to the thermal gradient. Obtained results are compared with the results due to Kosinski arid Hoffmann (2005). The experiments have shown that in some circumstances there is a low probability of an explosion propagating along the duct to produce an explosion in the second vessel. It appears that although the flame travels through the pipe and a jet flame enters the second vessel, it does not act as an effective ignition source. A local ignition can be quenched. The particles are here assumed to be ignited when their temperature exceeds a specified ignition value. So calculate the number of cells where the particle temperature is greater than 700 K and solid volume fraction is iii not less than 0.1, a User Defined Function (UDF) is developed. Greater the number of such cell, greater the probability of explosion. The presented results have been relatively simple and easy to interpret. The maximum error for temperature of the solid particles is ±30% and for solid volume fraction the maximum error is +15%. Further, effect of parameters like duct height, duct length, diameter of the particle, on the solid volume fraction and temperature of the gas have been discussed.