Concentrating solar power (CSP) plants with thermal energy storage offer several advantages to plants without storage. Thermal energy storage (TES) allows CSP plants to produce power for longer periods of time each day, making them produce energy more like traditional, fossil fuel power plants. TES also gives the ability to time shift production of energy to times of peak demand, allowing the plant to sell the energy when prices are highest. A CSP plant with storage can increase turbine performance and reach higher levels of efficiency by load leveling production and can remain productive through cloud transients. Power tower CSP plants are capable of producing extremely high temperatures, as they have the ability to oversize their solar field and achieve a greater concentration ratio. Studies have been conducted on variable working fluids, leading to higher working temperatures. This theoretically allows power towers to use more efficient, higher temperature cycles including the recuperated air Brayton cycle, although none currently exist on a commercial scale. This research focuses on developing a model of a high temperature TES system for use with an air Brayton cycle for a power tower CSP plant. In this research we model one module of a latent heat TES system designed to meet the thermal needs of a recuperated Brayton engine of 4.6 MWe capacity for six hours. A metal alloy, aluminum-silicide (AlSi), is considered as the phase change medium. The storage tank is approximately 161 m(3), or a cylinder with a 5 m diameter that is 8 m tall filled with AlSi with several air pipes throughout the volume. We model the volume around a single pipe in a 2-D cylindrical coordinate system, for a module size of 0.2 m in diameter and 8 m long. The advantages of using AlSi alloys is that they have variable melting temperatures depending on the relative concentration of the two metals, from 577 C for the eutectic composition of 12.6% silicon to 1411 C for 100% silicon. This attribute is taken advantage of by the TES model as the composition of the AlSi alloy will vary axially. This will allow a cascaded type storage system within one tank and with one material. The use of FLUENT to model this problem is first validated by several analytical solutions including Neumann's exact solution for a one-dimensional Cartesian geometry and the Quasi-Steady Approximation in a 1-D cylindrical geometry. The model developed will establish charge/discharge times for the storage system, round-trip efficiency of the system, ability of the system to meet the demand of the Brayton cycle, and the. validity of using off-eutectic metal alloys in a cascade as a latent heat TES medium.
