A STOCHASTIC-PROBABILISTIC MODEL FOR SIMULATING PARTICLE DISPERSION IN GENERAL COORDINATES

Abstract

This article describes an extended work on the development of the stochastic-probabilistic particle dispersion model for turbulent two-phase flows in general coordinates. The extended stochastic-probabilistic modeling of particle dispersion makes it possible to achieve a smooth particle flow field by tracking a small number of particle trajectories. A numerical method was developed for determining the spatial distribution of physical particles in irregular Eulerian control volumes. As a result, the extended particle dispersion model can be used, together with the Eulerian modeling of fluid flow, to predict turbulent two-phase flows with complex geometry. Compared with the conventional stochastic model, the present stochastic-probabilistic model has overcome the deficiency of tracking too large a number of particle trajectories to achieve a smooth particle flow field. Numerical results were reported for a particle-laden turbulent gas flow and liquid flow with available experimental measurements. The performance of the present model was assessed in terms of its agreement with experimental measurements and its computational efficiency as compared to the conventional stochastic particle dispersion model. For the two test cases considered, it was found that computational efficiency has been enhanced by 75%.