An efficient method for modelling thermal energy storage in packed beds of spherically encapsulated phase change material
Colin Jaffray, Kamran Siddiqui, Anthony G. Straatman
Abstract
An approach for modelling melting (and solidification) in packed beds of encapsulated spherical PCM is presented. The approach includes a calibration step based on comparisons of simulations with experimental results of PCM melting in a cylindrical geometry, followed by detailed simulations of melting in an isolated encapsulated PCM sphere and a PCM sphere in a representative elemental volume of packed PCM spheres to study the impact of external flow conditions on PCM heating and melting. The detailed simulations show that the enthalpy-porosity model provides a reasonable estimate of the overall melting time and a good approximation of the evolution of the liquid/solid melt front during the melting process. The simulations also confirm the dominance of convection during most of the melting process. The temporal integral results of overall heat transfer coefficient and overall energy fraction from the detailed simulations are combined to yield a relationship between overall heat transfer coefficient and overall energy fraction such that the process can be modelled in terms of the condition inside the encapsulated PCM sphere, wherein the condition can also be obtained from the fraction of heat absorbed compared to the energy capacity of the PCM sphere. This relationship is then used as an input to a finite-volume model for packed beds which can be used to predict the charging time of entire packed beds of encapsulated spherical PCM. Results from the simple model show correct trends for the temporal evolution of liquid fraction and PCM temperature, and for overall charging times for the cases considered. This simple approach can substantially reduce the simulation time required to model beds of different shaped encapsulated PCMs and different dimensions. • Thermal expansion coefficient is shown to be a suitable calibration parameter for PCM melting. • Detailed simulations of PCM encapsulated sphere inside REV used to mimic packed bed • Average heat transfer coefficient as function of energy fraction can be used to characterize heating process. • Finite-volume model can utilize output from single sphere to simulate charging of packed bed.