Optimal Parameter and Fin Design for PCM-based Thermal Energy Storage

Latent heat storages (LHSs) have gained a lot of interest due to their high potential of creating compact and efficient thermal energy storages, which are highly desirable in our current and future energy market for increased use and integration of renewable energy systems and to pursue carbon neutrality. However, charging of latent heat storages is still hampered by the low thermal conductivity of the Phase Change Material (PCM). This greatly limits the achievable powers and steeply rises the time for charging and discharging the storage. Moreover, the limited power increases the temperature difference between the heat transfer fluid (HTF) supply temperature during charging and the HTF output temperature during discharging, resulting in losses in potential of the supplied thermal energy. In order for LHSs to become competitive in the current energy market, their charging performance has to be improved. Current heat transfer enhancement technologies, such as highly conductive fins and PCM encapsulation, have proven to be valuable tools, but are still lacking general design rules. In this work, a numerical model is presented for calculating the thermal response of LHSs and their according charging and storage performance. In view of LHSs based on low temperature differences and small PCM enclosures, the energy model is reduced to a transient heat conduction and phase change model, neglecting the effect of natural convection. The modelling along with a dimensionless analysis is used to investigate the effect of different PCM shapes through encapsulation, the heat transfer enhancement by highly conductive fins, and the effect of a HTF with a limited input power. Moreover, optimization strategies are envisaged for improved LHS design. The combined charging and storage performance of four different PCM shapes is investigated: cuboid shaped PCM, cylindrical PCM, spherical PCM, and PCM in tubes. The charging trends for the different PCM shapes appear to be very similar. The results show that for all shapes, reducing the size of the PCM enclosures greatly enhances the charging performance. Moreover, the PCM shape greatly influences the maximum achievable storage density, due to intermediate voids. Due to the similarity between the charging trends of the different shapes, only cuboid shaped PCM is selected to be studied more in depth, using uniformly distributed, highly conductive fins and by including the HTF model for a hot water flow which acts as the heat source. A combined numerical-analytical study reveals the limits of PCM charging using uniformly distributed fins, which are achieved by designing infinitely thin fins. A Pareto front between energy and power density is obtained, presenting the trade-off between storage capacity and charging rate for these idealised limits. Further, the performance of realistic metal fins is investigated, which presents the lowered charging performance due to the finite fin size, however, the performance does converge to the charging limits for thinner fins. Restricting the input power of the water flow results in a non-uniform melting of the PCM, due to water temperature variations. Therefore, latent heat storage tanks are also modelled with a water flow with a finite total heat capacity. The results show that increased charging ratios, i.e. the ratio of stored heat to input heat, can be obtained for high flow lengths, reduced PCM layer thicknesses, and reduced input power. Additionally, both storage and charging performance increase for a lowered water channel height. Further, we show that designing the storage as a stacking of multiple parallel chargeable PCM compartments enables the charging performance to be improved and the pumping power requirements to be reduced. Finally, optimal fin distributions and optimal topological fin designs are presented for latent heat storages charged by a constant input power water flow. As the limited input power results in a non-uniform melting of the PCM, new LHS designs with non-uniform fin designs provide the opportunity to outperform the ones with a uniform shape and/or distribution. We show that different optimal distributions of rectangular fins are found depending on the fin width, amount of fins and the input power. Also optimal fin distributions for increased PCM compartment lengths are shown to be very similar. Moreover, new fin topologies are obtained using density based topology optimization. The fin designs are shown to be tree-like and heavily branched for high power requirements. The gain in charging performance is discussed by comparing the optimized designs with latent heat storages with uniform fin distributions and show charging power increases up to 5.5%.status: publishe