Modeling the change in particle size distribution in a gas-solid fluidized bed due to particle attrition using a hybrid artificial neural network-genetic algorithm approach

Abstract

Particle size distribution (PSD) is an important parameter in gas-solid fluidized bed. The change in PSD due to particle attrition can affect the long-term performance of fluidized bed. In this study, artificial neural network (ANN) with genetic algorithm (GA) as a meta-modeling tool was employed to model the change in PSD during fluidization. Experiments were conducted using incineration bottom ash (IBA) as the fluidizing particles and different mass percentage of large and small glass beads were used as the grinding medium. Rosin–Rammler (RR) distribution was used to describe the IBA PSD. The ANN-GA models developed were subsequently used to study the effect of fluidization time, mass percentage of glass beads and size of glass beads used on the IBA particle attrition during fluidization.