Paper presented to the 3rd Southern African Solar Energy Conference, South Africa, 11-13 May, 2015.
Packed beds incorporating rock as the storage medium and air as the heat transfer fluid have been proposed as a cost-effective approach for thermal storage in solar power plants. In order to assess the viability of rock bed thermal energy storage (TES), it is essentially necessary to predict the air and rock temperature profiles through the bed during charging and discharging. A number of models are available for this purpose, including, at the more basic level, the popular Hughes “E-NTU” model and the simplified “Infinite NTU” model. Unlike the E-NTU method, the Infinite NTU method assumes an infinite heat transfer coefficient between the air and the rock, which implies equivalent air and rock temperature profiles. Typically, the more detailed a model, the greater the computational effort required to solve it. For long-term analysis, a time-efficient model is necessary to prevent excessively long computation times. This paper evaluates the comparative accuracy and computational efficiency of the E-NTU and Infinite NTU models when simulating the short- and long-term performance of CSP rock bed TES systems. On the basis of these comparisons, the appropriateness of employing the less realistic but less costly Infinite NTU model in the long-term simulation of CSP rock bed TES systems is evaluated.