ABSTRACT
An efficient algorithm for particle-particle and particle-wall collision detection in the two-dimensional case is briefly presented. The algorithm ensures an efficient computation of colliding particle flows. The physical domain is hierarchically divided and structured as a quad-tree. The algorithm is intended for particle-laden flows, which require small time steps, but within each time step, the algorithm is event-driven. The algorithm is used here to study the flow behaviour of a laboratory-scale fluidised bed. The two-dimensional motion of each individual spherical particle is directed, calculated from the forces acting on it, accounting for the interaction between the particle and the gas phase. The soft sphere model is used in the present work. The contact forces are continuous and finite and are a function of the deformations of the particles. A comparison of experimental observations and computational results of a twodimensional laboratory-scale bubbling fluidised bed is presented and discussed.