Numerical Investigation of Weather-Driven Transient Thermal Behavior in Direct Steam Generation External Receivers of Solar Power Tower Plants
Loading...
Files
Size: 8.52 MB, File format: Adobe PDF
Date
Authors
Al-Sarraf, Hayder
Supervisor
Zamora, Ramon
Alhusseny, Ahmed
Item type
Degree name
Journal Title
Journal ISSN
Volume Title
Publisher
Auckland University of Technology
Abstract
Solar power tower plants are promising candidates to decarbonize electric power production. These plants concentrate solar thermal power to raise the temperature of the heat transfer fluid. However, due to atmospheric effects and cloud events, this thermal power intensity varies spatially and temporally throughout the diurnal cycle. This PhD thesis deeply investigates the effect of weather conditions variability: solar irradiance, ambient temperature, and sky temperature, on the thermal performance of water-cooled external receivers employed in solar power tower plants. First, this study identifies knowledge gaps in the literature regarding the evaluation of solar thermal power incident on receiver panels. Evaluating the net solar thermal power delivered to the receiver tubes as a function of time and position is therefore essential. Moreover, this study introduces a comprehensive approach that enables the derivation of a time-dependent heat-flux equation for external receiver-absorbing tubes in the Ivanpah I plant using the SolarPILOT tool. In addition, a modified Gaussian distribution is derived for the incident heat flux over the tube circumference. Compared to the proposed distribution, both the uniform and the basic Gaussian distributions employed in former computational fluid dynamics simulations would result in about 57.1% overestimation of the total solar power received. To achieve high-accuracy correlations in the temporal heat irradiance equation, multisegment correlations are established for the temporal profiles of the axisymmetric heat flux on each side of the receiver. These correlations fit the data well; the minimum R2 values are 0.9989, 0.9989, 0.9994, 0.9985, and 0.9988 for the multi-segment correlations generated for the northern, eastern, southern, and western sides of the receiver, respectively. In this research, a comprehensive thermodynamic and thermal analysis procedure is developed and applied under real-world weather conditions to demonstrate the potential of the proposed scheme to perform such complex computations cost-effectively. Based on the proposed procedure, an in-house MATLAB code is developed to numerically predict instantaneous heat losses, net heat flux, steam bulk temperature, tube wall temperature, and vapor fraction for both the evaporator and superheater sides. The results reveal that the onset of nucleate boiling in the north-facing evaporator tube takes up to 2 hours from sunrise to reach, with 70% of the tube length required to initiate evaporation, which then continues for the rest of the tube. However, superheating can be established once solar intensity is strong enough around midday, occupying up to 12.9% of the tube length. From an operational point of view, three scenarios are discussed regarding their impact on steam bulk temperature, productivity, and enthalpy. Among them, Scenario #3 outperforms in terms of net productivity due to the lower overall makeup required throughout the day, with the receiver meeting 93.61% of the plant steam demand in standalone mode, compared to 90.44% and 89.06% when Scenarios #1 and #2 are followed, respectively. From a safety perspective, the wall temperatures of the superheater tubes on the north, east, and west sides exceed the maximum allowable limit. To address this issue, a mass flow interchange approach with circulation factors between the opposing sides is proposed using a temperature control valve. It was found that the uneven distribution of steam fed to the superheater sides not only ensures the receiver's safety but also slightly reduces the total makeup required while increasing the excess energy available. It results in a 0.75% reduction in the energy required to meet the turbine inlet condition, with a 0.626% reduction in the net makeup ratio of the receiver. Eventually, the current research could assist in future detailed CFD investigations of external solar power receivers and is significant for ensuring the reliability and longevity of such systems. To ensure both system reliability and safety, a series of optimization procedures has been performed to identify optimal circulation factors for each opposing pair of superheater panels. The optimization reveals that the circulation factor is 0.25 from south to north and 0.1375 between east and west panels. Adopting this approach and the optimal circulation factors not only ensures the wall temperature remains below the maximum allowable limit but also results in a marginal improvement in energy utilization and conversion. As a result, the optimized operation reduces the net power required by 0.7731% and the makeup ratio by 0.782%. Moreover, the electricity output under the optimized approach matches that of the non-optimized scenario. Moreover, sensitivity analyses of the approximation scheme adopted are conducted to assess the thermal performance of the solar receiver under several influential factors. Results show that high-resolution solar irradiance (10-min DNI) is sufficient for evaluating receiver performance, compared with 1-min DNI. Hence, the error in computing total solar energy density varies marginally in the morning and afternoon periods, i.e., 0.183% and 0.06%, respectively. In addition, estimating the thermal conductivity and dynamic viscosity of steam at the instantaneous bulk temperature does not significantly affect receiver performance. Hence, the errors introduced in the total makeup power, excess power, and makeup ratio are 0.01%, 0.3%, and 0.4%, respectively, compared to constant magnitudes for these properties. Also, a comparison between different clear-sky temperature models is carried out. Even though Model#3 provides higher sky temperatures, which should consequently yield lower radiative losses, it shows higher total heat losses than Model#1 due to increased convective thermal losses. Therefore, model selection is highly dependent on the plant's actual geographical and meteorological conditions. Finally, the impact of wind speed and direction on the receiver's convective losses is evaluated. Results show that parallel-flow wind causes greater convective heat loss than cross-flow wind, in good agreement with the literature. The total makeup power and excess power increase by (4.55%) and (14.74%) when the wind direction is parallel compared to cross-flow. Therefore, it is recommended to consider wind speed as an influential factor in predicting the thermal performance of the solar receiver.