Estimation of wind erosion emission from iron ore piles using CFD and Von Karman methods

Open aggregate storage piles are used more and more in industrial sites. In industrial areas, emitted particulate matters from piles of row material can affect the quality of life of workers and employees and also the quality of the environment. Study of dispersion patterns and concentration of particulate matters over a landscape is important for the strategy of monitoring and controlling particulate matter. Within an industrial facility, dust emission may be generated by wind erosion of open aggregate storage piles and therefore, it pollutes the environment and wastes the row materials. Emission of particulate matters from surface of a pile depends on many parameters such as characteristics of wind (e.g. wind speed and wind direction), specifications of particles (e.g. particle diameter, density, shape, etc.) and erosion properties of surface. Therefore, for emission calculation of particulate matters from a pile and also for simulation of dispersion of emitted particles, it is necessary to simplify the physics of this phenomenon. Simplifications have been carried out based on governing equations and also applying the empirical relations obtained by field studies. Based on these theoretical and empirical investigations, a few methodologies are available for atmospheric wind erosion calculations from storage piles of row materials. The U.S. Environmental Protection Agency (EPA) method is one of the most famous approaches to this kind of calculation. It focuses on estimating the wind erosion and it cannot be used for dispersion pattern prediction. On the other hand, some models and methods have been developed to calculate the dispersion of pollutant in near and far distances from sources. One can combine the calculation of particulate matters emission with dispersion models in order to determine the particulate matters concentration at the environment. In the present work, two methods including the U.S. EPA wind erosion estimating approach combined with Von-Karman's scheme for dust settlement and computational fluid dynamic (CFD) method using Fluent 6.3.2 Software are applied to predict particulate matters dispersion patterns from an iron ore pile. The Von-Karman's method is based on the length and time of the particulate matter settling. In the present work, the concentration of particulate maters in different distances from the source has been calculated using these parameters. In the CFD technique, the geometry of a pile is generated in Gambit Software using a structured mesh tool. The number of the generated mesh on the pile is 104,214. In this study, the flow condition is assumed to be incompressible, turbulent and steady state. Turbulence modeling is carried out based on two types of modeling namely k  and k  theories. Atmospheric wind profile is assumed to be in neutral conditions and defined by a user-defined function (UDF) tool from Fluent Software. The results from the two methods are compared with concentration of particulate matters measured based on 10 points. The maximum concentration position predicted by the CFD approach is more precise than that predicted by Von-Karman's method. Good quantitative and qualitative agreements are observed between the CFD predicted deposition and the measurement results. The determination coefficient for CFD and Von-Karman methods are 0.71 and 0.35, respectively. Also, Von-Karman method underestimates the concentration of particulate matters in all 10 measurement points.