An air-rock bed thermal storage system was designed for small-scale powered generation and analyzed with computational fluid dynamics (CFD) using ANSYS-Fluent simulation. An experimental system was constructed to compare and validate the simulation model results. The storage unit is a cylindrical steel container with granite rock pebbles as a storage medium. The CFD simulation used a porous flow model. Transient-state simulations were performed on a 2D axisymmetric model using a pressure-based solver. During charging, heat input that keeps the bottom temperature at 550 °C was applied to raise the storage temperature. Performance analysis was conducted under various porosities, considering natural and forced convection. The natural convection analysis showed insignificant convection contribution after 10 h of charging, as observed in both average air velocity and the temperature profile plots. The temperature distribution profiles at various positions for both convection modes showed good agreement between the simulation and experimental results. Additionally, both cases exhibited similar temperature growth trends, further validating the models. Forced convection reduced the charging time from 60 h to 5 h to store 70 MJ of energy at a porosity of 0.4, compared to natural convection, which stored only 50 MJ in the same time. This extended charging period was attributed to poor natural convective heat transfer, indicating that relying solely on natural convection for thermal energy storage under the given conditions is not practical. Using a small fan to enhance heat transfer, forced convection is a more practical method for charging the system, making it suitable for power generation applications.