Assessment of the Predictive Capability of Eulerian-Eulerian Modeling Approach for the Spouted Bed Thermal Receivers
Özet
Directly irradiated fluidized/spouted particle receivers can potentially store thermal energy at higher temperatures than 1100 °C in concentrated solar power (CSP) applications. Therefore, these types of thermal receivers are good candidates for the next-generation CSPs. Furthermore, particles suggested for CSP applications such as carboHSP and olivine have high density (larger than 2500 kg/m3) and for particles sizes of 1 mm or more, spouted beds are usually preferred over fluidized beds due their low bed pressure drop characteristics.
The number of thermal numerical studies on the spouted beds for CSP applications are limited in the literature. One commonly used modeling approach is the Eulerian-Eulerian approach (also known as TFM) where both solid phases is defined as a continuum similar to gas phase. Most TFM studies in the spouted literature investigates the gas-solid flow dynamics whereas available thermal studies do not address the gas-solid flow dynamics in detail. Furthermore, TFM approach needs various constitutive equations and adjustment of hydrodynamic and thermal properties for which a comparative assessment is needed to develop reliable models. Therefore, the aim of this thesis is to investigate the prediction performance of TFM for the CSP thermal storage applications in spouted beds assessing the hydrodynamic and thermal performance simultaneously. The scope of the thesis is limited to discharge behaviour of the thermal storage system by convective cooling. The radiative heating has not been considered. In this context, there different data sets were selected (He et al. (1994a; 1994b), Patil et al. (2015) and Ozdemir (2024)) for validation. While He et al.’s (1994a; 1994b) data set is strictly for gas-solid flow dynamics, Patil et al. (2015) and Ozdemir (2024) provide thermal data sets for convective cooling simulating the discharge behaviour in CSP systems. For the simulations, ANSYS FLUENT was used and interphase heat transfer and different effective thermal conductivity models were implemented using user defined functions.
The results indicate that amongst the hydrodynamic TFM parameters the drag model and the restitution coefficient are the most critical ones from hydrodynamics perspective to obtain reasonably accurate solutions. On the thermal side, the gas-solid interphase heat transfer coefficient, the effective thermal conductivity of the solid phase, and the appropriate definition of the wall boundary conditions are essential for accurate modeling of the cooling or thermal energy discharge behavior of the bed. The correlation of Gunn (1978) for the interphase heat transfer coefficient result in particle Nusselt numbers (Nup) consistent with the measurements in the fluidized bed literature and the model of Zehner and Schlunder (1970) for the solids phase effective conductivity perform better than the kinetic theory approach. The energy balance and thermal mesh sensitivity are two important points which are often overlooked in the literature. Obtaining a mesh sensitive solution based on hydrodynamic parameters does not guarantee mesh sensitivity for thermal simulations. Thermal simulations require finer mesh sizes. Finally, comparison of the results with the Ozdemir’s (2024) data set for the discharge behaviour of carboHSP particles indicate that the TFM can capture the general trends of the cooling behaviour for the first 100 s for which the simulations were run. The maximum temperature difference between the simulation and experimental results based on the data obtained from three locations in the bed is 7-8 °C and the cooling rates are reasonably similar. Furthermore, TFM simulations provide some physical insight on the cooling behaviour of the CarboHSP particles.
