Numerical modeling of the flow of organic fertilizers in land application equipment

The knowledge of the flow behaviours of organic fertilizers in land application equipment as well as the machine-product interactions is very sparse and empirical and can hardly contribute to the design of innovative high-performance systems or to the optimization of the operational parameters of available machines. To assess the potential of using numerical modeling as a tool to optimize the design and operation of land application equipment, the specific objectives of the work reported herein were to model the flow of organic fertilizers in such machines and to validate the models against various parameters measured during field experiments using commercial spreaders. The discrete element method (DEM) was used to simulate the flow of compost while computational fluid dynamics (CFD) was applied to sludge spreading. Input parameters for both modeling methods were derived from physical and theological properties of composts and sludges. The operation of a spreader featuring dual vertical beaters was first simulated and the ground distributions and power requirements obtained were consistent with published results for that type of machine. Two types of composts were modeled in simulations aimed at studying the effect of a flow-control gate on the discharge flow and energy requirements of the discharge conveyor. Dimensionless parameters were developed to allow for comparisons to be made between the scaled simulations and the experimental results obtained with a full size spreader. Simulated results for the discharge flow in the open gate configuration were in good agreement with measured data. The mass efficiency values for dry compost, which represent a dimensionless measure of the discharge rate, were 1.21 and 1.22 for the measured and simulated cases, respectively. The effect of the gate on the power requirements of the discharge conveyor were replicated by the models. Fracture behaviours observed in the bulk of product during field experiments were also replicated by the model. Two types of sludge exhibiting different theological properties were modeled using a CFD code. Field experiments were carried out with a sludge spreader to measure the discharge rate of the spreader for these types of sludges. The simulated flow rate curves closely replicated the experimental ones, for both sludge types. The simulated streamlines during the unloading of the spreader were also well correlated to observations made during field experiments. The flow of sludge on a spinning disc was studied using high-speed photography and a scaled spreader physical model. The velocity of the sludge on the disc could be calculated using two different parameters measured on the images. It was found that the viscosity of the sludge influenced the spreading pattern. The flow of sludge on the disc was also numerically simulated with CFD. The simulated and measured residence time of the sludge on the disc were influenced by the viscosity of the sludge and were in close agreement. The measured and simulated residence times for fluid pasty sludge were 0.021 and 0.025 s, respectively. The measured and simulated values for plastic pasty sludge were 0.037 and 0.033 s, respectively. The DEM was successfully applied to the flow of compost and CFD was effectively used to model the flow of sludge in a land application machine. Both numerical modeling techniques showed promising results and have the potential to become more accurate through more detailed simulations and improvements in the products constitutive models. (c) 2005 Elsevier B.V. All rights reserved.