This thesis consists of two main parts. In the first part, we aim to investigate the segregation and mixing behavior of non-spherical Geldart B and D E-CAT particle mixtures in a bubbling fluidized bed. The experiments will focus on both unsieved and sieved E-CAT particles, with varying particle size distributions and fluidization velocities ranging from 1.3 to 5Umf. High-speed pressure and image data will be utilized to quantify the dynamic behavior of the bed, including the expanded bed height and pressure drop. Additionally, we will develop a hybrid machine learning-aided image processing algorithm for quantifying the instantaneous particle size distribution at different axial positions. The goal is to explore how segregation behavior correlates with particle size distribution and fluidization velocity and to gain a deeper understanding of the underlying processes, which will aid in the optimization of fluidized bed design and numerical model validation.
In the second part, we propose a novel UAV-based platform for real-time measurement of wind and particle distributions within the atmospheric boundary layer. The system will integrate a full-scale sonic anemometer with a multi-resolution digital in-line holography (MRDIH) system to capture three-dimensional wind vectors and particle size distributions, ranging from micrometer-scale aerosols to millimeter-sized particles. Time-domain and frequency-domain correction methods will be employed to mitigate rotor-induced disturbances and ensure accurate wind measurements. This platform will provide valuable insights into wind–particle interactions, including droplet dynamics in clouds and turbulence-driven particle dispersion, with applications in wind energy assessment, air-quality monitoring, and research in multiphase flow dynamics.
