Particle Movement and Separation Phenomena for a Gravity Separator: II. Experimental Data and Performance of Distance-Transition Markov Models.

Abstract

Linear programming procedures are used to fit stationary and nonstationary distance-transition Markov probability models to experimentally obtained particle distribution data from the deck of a gravity separator. Particle movement modeled is that of a light discard fraction. Performance of the models is examined by comparing predicted "typical'' particle pathways with observed particle pathways. The Markov process is found to be an appropriate model of particle movement.