This presentation was made at the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

The way particles arrange themselves in a storage silo can influence airflow paths and airflow distribution as a result. That relationship is of interest to the agriculture industry where quality preservation of stored grain crops often requires aeration or drying with forced air. Other industries also need to store bulk materials and the properties of the bulk material can have an influence on various processes that occur during or after storage. The voidage gradients that develop inside grain bulks was investigated for various filling methods and silo geometries. The discrete element method (DEM) was used to model wheat kernels stored in scaled silos. Discrete values of local voidage from the DEM simulations were mapped using a Delaunay transformation. Distributed (rainfall) filling generated higher packing densities than more concentrated central or eccentric filling. The impact of the location of filling with respect to the center of the silo had limited impact on the radial distribution of voidage. Higher voidage values were observed in the bottom portion of hopper-bottom silos compared to flat-bottom silos under the same conditions. The presence of a physical obstruction like an aeration device placed inside the silo resulted in relatively complex voidage profiles. Lowering the shear modulus of the wheat particles to reduce the computational cost of the simulations was found to affect the results but did not invalidate the observed voidage trends.



A number of questions remain regarding the consequences of the observed trends in voidage gradients. The biggest challenge of the analysis presented is the scalability of the results. Real-world bulk material systems include large numbers of particles that have always been a challenge for discrete element modeling. Opportunities exist to investigate scaling laws for the observed behaviours and to link the discrete element approach to a fluid dynamics method focusing on porous media properties.