| Turbulent particle dispersion is a phenomenon that is of significant practical importance to coal combustion as well as other disciplines. The lack of an efficient and accurate description, of the interactions of the condensed phase with the turbulent, reacting environment, is a major obstacle in the development of computational simulations of these important industrial processes.; Turbulent particle dispersion modeling is complicated, not only by a lack of understanding of the turbulent flow field itself, but also by the inherent complexity of coupling fluid and particle dynamics calculations. Traditional statistical Lagrangian methods, at best, use Monte Carlo simulation techniques. This entails tracking the history of a large number of particle trajectories to achieve stable and accurate solutions, making them too computationally intensive to be used in simulations of practical industrial applications. The conceptual approach used herein involves tracking the statistical evolution of clouds of particles through the computational flow domain. The distribution of particles within the cloud is represented by a Gaussian probability density function (PDF). The mean of the PDF for particle position is obtained by solving an ensemble averaged equation of motion. The variance of the PDF is a measure of the turbulent particle dispersion and is modeled by the particle velocity correlation function. Correlations, derived by modeling the turbulent transport mechanism of the particles on a fundamental basis, have been used to provide a physically realistic description of the turbulent particle dispersion phenomenon.; The particle cloud PDF tracking approach provides for efficient and accurate transport, reaction and radiation coupling with the continuous phase calculations. The submodel has been implemented within the framework of a three-dimensional computational simulation for gaseous reacting flows and tested by performing simulations of several cases. The dispersion submodel itself has been validated by comparison of simulation predictions with reliable experimental dispersion data. Coal combustion reactions have been interfaced with the particle phase calculations to provide the capability of simulating coal combustion applications. The performance of this overall simulation has been demonstrated by simulating the experimental pulverized coal-fired furnace at the International Flame Research Foundation (IFRF). A third case study of the process of sulphur dioxide capture in flue gas by sorbent injection demonstrates the general applicability of the simulation scheme and its usefulness as an engineering tool to infer causal relationships in complex processes.; Thus, a computer program capable of performing accurate and computationally efficient three-dimensional predictive simulations, of industrial processes involving turbulent particle dispersion, such as large-scale coal combustion furnaces, has been developed, validated and demonstrated. |
