Experiment and numerical simulation were carried out in this work to investigate the thermal transport behaviors of Newtonian fluids, phase change material (PCM), and composite PCM in a cavity, filled with porous media for thermal energy storage (TES) applications. Herein a novel experimental system was developed by using numerous long and parallel steel wires embedded in the cavity to construct a two-dimensional (2-D) porous structure, which provided an accessible way for numerical simulation by directly meshing the porous structure. As Rayleigh number (Ra) over than 1.0E8, the flow was simulated by standard k-epsilon turbulent model. The effects of heat conduction through metal structure, natural convection of liquid PCM, and geometric configuration of porous medium and porosity on improving the thermal transport efficiency were discussed. For Newtonian fluid flow and heat transfer, higher Rayleigh number led to stronger natural convection, and the deviation between experimental and numerical is less than 10% that confirmed the reliability of numerical method for further application in phase change process; for phase change behavior of PCM in the cavity with wires, natural convection was apparently weakened as porosity (epsilon) decreased to 80% and heat conduction started to dominate the heat transfer. The full charging time Wait wire) of PCM mixed with wires was then compared to that (t(full, EG)) of PCM mixed with expanded graphite (EG) under the same porosity. The results showed that EG significantly improved the heat transfer efficiency that t(full,EG) for four cases (epsilon = 95%, 90%, 85%, 80%) is only 24.4%, 9.5%, 5.7% and 4.0% of t(full,wire), respectively. This result indicates that when mixing PCM with material of high thermal conductivity to improve the thermal conductivity of composite PCM, the porosity and geometric configuration of the material are essentially needed to be considered. Furthermore, the cross-linked structure of material plays a more important role than porosity in enhancing thermal transport efficiency through heat conduction. This work brings deep insights into sensibly improving the thermal conductivity of PCM by adding materials with high thermal conductivity.