Titolo della tesi: Heat transfer models of a hierarchical high-temperature solar receiver for solar heated industrial processes
Among renewable energy sources, solar energy has the potential to supply a big part of the necessary heat of the industrial sector to pursue the EU Green New Deal climate neutrality purpose. Concentrated solar power technologies use mirrors to concentrate the solar radiation and produce heat. Point focusing configurations, like central receiver towers or solar dishes, are capable of reaching high concentrated radiative fluxes over 500kWm-2, which can produce high-temperature heat over 700°C. In concentrated solar power plants, solar receivers are one of the key components, which convert the concentrated radiation into usable heat. A particular type of solar receiver, the so-called volumetric, utilizes a porous structure to absorb the radiation in the whole volume. In the meantime, the heat is transferred by convection to a fluid that is forced to flow through the same structure. In these structures, highly efficient conversion from radiation to heat is crucial achieving high temperatures and unveiling the possibility to support high-temperature industrial processes.
The metallic hierarchical volumetric receiver (developed in FBK), which presents intricated highly ordered geometry, has demonstrated a 72% thermal conversion efficiency but lower than expected. The proper design of the absorber has to combine mainly two heat transfer mechanisms, convective and radiative, finding a trade-off between geometrical parameters like porosity, wetted area, material surface characteristics and optical properties. Therefore, this work presents a methodology, including models and experiments, to understand and quantify the heat transfer phenomenon inside the hierarchical absorber.
The problem is, thus, divided into three main parts: characterization of radiative heat transfer, analysis of the heat exchanged by convection, and experimental tests of two upscaled prototypes in a high flux solar simulator. The radiative heat transfer is widely investigated, starting from the analysis of a real solar dish, where Monte Carlo ray-tracing simulations are compared with experimental data. The radiation propagation inside the hierarchical structure is examined in both experimental and numerical methods. A self-developed instrument based on single-photon avalanche diode sensors is employed to characterize the optical behavior of the hierarchical absorber. Ray-tracing techniques are furtherly employed to analyze the absorber in both the radiation propagation and infrared emission by means of configuration factors. In order to assess the emitted energy to the environment, a new parameter is defined as a surface averaged view factor. The convective performance of the hierarchical geometry is examined in computational fluid-dynamic simulations and compared to a reference absorber. Furthermore, a high-flux solar simulator is utilized to characterize the performance at several working conditions of two upscaled prototypes. One of them is a new structure, resulted from a dedicated design process, presenting a thicker geometry of the absorber.
As a result of this work, several reliable models are developed investigating in detail the hierarchical absorber behavior. From these models, an optical “trap effect” for the investigated absorber is discovered, which means that a re-absorption of the reflected light from the internal part of the absorber occurs in the frontal layers. The two examined geometries present a high optical efficiency, defined as the ratio between absorbed and incident power. The new longer geometry shows a 5% performance increase, with 90% of absorbed radiation. In the infrared emission analysis, the hierarchical absorber directs most of the radiation emitted to itself. The newly introduced parameter shows that 11% of emitted power is directed to the surroundings. The hierarchical absorber Nusselt number reaches high values in the layer interfaces, up to 14. However, in other zones, the considered receiver presents lower performance than the considered reference. In high-temperature experimental tests, the new geometry seems to perform better than the original, resulting in an average efficiency increase of about 20%. The efficiency reaches a maximum of 92.9% at 107 kJ kg-1, which is decreased to 53% when the power to mass ratio is increased.
As a consequence, the new proposed geometry presents an enhancement of hierarchical structure, unveiling the reliability of the methodology developed herein. The receiver can be further optimized by exploring the effect on the efficiency of parameters such as the receiver length. Further optimization might be achieved by employing in new designs the model ecosystem here illustrated, pushing the maximum air temperature to further increase.