EFFICIENT COMPUTATION OF PARTICLE DISPERSION IN TURBULENT FLOWS WITH A STOCHASTIC-PROBABILISTIC MODEL

This paper describes a three-dimensional, stochastic-probabilistic, ef ficiency-enhanced dispersion (SPEED) model for the prediction of parti cle dispersion in turbulent hows. A stochastic procedure is used to co mpute particle-trajectory mean and variance, and a probabilistic proce dure is used to compute a physical-particle spatial distribution among Eulerian control volumes. An additional Lagrangian equation is derive d to govern the evolution of particle-trajectory variance. The SPEED m odel is aimed al tracking a relatively small number of particle trajec tories, while efficiently reducing computational shot noise in the con ventional stochastic dispersion model. Two tests cases with available measurements are used to validate the efficiency of the SPEED model. N umerical results of the SPEED model using only 5 x 10(2) particle traj ectories are compared with those of the conventional stochastic model using as high as 1.5 x 10(4) particle trajectories. It is found that t he SPEED model offers better agreement with the experimental measureme nts, and that the computational efficiency can be substantially enhanc ed by a Factor of 20. (C) 1997 Elsevier Science Ltd.