Date of Award
Spring 2024
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Mechanical Engineering
First Advisor
Wilbert Cruz
Abstract
In this study, a 3D steady-state Eulerian-Eulerian simulation was developed to investigate flow performance in a dilute pneumatic conveying system. The investigation focused on two main objectives: (1) assessing the influence of bin selection on solution accuracy for representing the polydisperse mixture, and (2) exploring methods to simplify the modeling of polydisperse particles in the simulation by utilizing a mean diameter derived from the mixture’s particle size distribution. The system's geometry and initial boundary conditions featured a 10.6 m long, 150 mm diameter horizontal pipe with both gaseous and solid phases initially moving at 27 m/s . The steady-state simulation replicated an experimental pneumatic conveying system built by Laín and Sommerfeld, to compare and validate the proposed Eulerian-Eulerian model with experimental data. The dilute mixture used in this study mirrored Laín and Sommerfeld’s particle size distribution (PSD), consisting of particle diameter sizes of approximately 15 μm to 85 μm. By incorporating an inhomogeneous discrete population model, a set of five bin cases were conducted using Laín’s PSD in the Eulerian-Eulerian simulation. An additional monodisperse case was examined, featuring particle sizes of 40 μm which corresponded to Laín and Sommerfeld's PSD mean diameter. This comparison aimed to evaluate the efficacy of representing the polydisperse mixture via bin characterization techniques, as opposed to utilizing a single mean diameter from the PSD. After validating the model, a series of four PSDs, ranging from 5 μm to 500 μm, were examined. Bin convergence studies were conducted for each PSD, as well as the viability of utilizing mean diameter expressions to represent each particle size distribution was investigated. Through this study, it was found that approximately 20 bins were necessary to accurately capture the pressure profiles of the system for most PSDs. Furthermore, simulations utilizing particle sizes based on the Sauter mean diameters demonstrated the most favorable agreement in pressure drop predictions, showcasing up to 0.086% deviations with highest bin cases. The remaining mean diameter expressions gave rise to similar system pressure drop values, with De Brouckere being the second most accurate, and volume mean, surface area mean and arithmetic mean diameters following respectively.