Creating dimensionless performance curves for latent heat thermal energy storage

Latent heat thermal energy storage (LHTES) systems can be used to reduce electricity demand when used in conjunction with combined heat and power plants or HVAC (Heating, Ventilation, Refrigeration and Air-Conditioning), as they can regulate the demand and supply of thermal energy. LHTES can also be used to integrate renewable energy sources with the grid. A design procedure and performance modeling is imperative for the effective utilization of thermal energy storage in HVAC equipment and systems. Current LHTES models utilize empirical performance curves that are based on experimental results or high fidelity simulations. While such empirical performance curves for LHTES may be utilized for modeling their respective design configuration and conditions, their dependence on operating conditions and geometry limit their applicability. We present a non-dimensionalized approach based on flow physics to create a versatile LHTES performance evaluation metric which is agnostic of the specific operating conditions or geometry, but depends on the flow regime observed in the LHTES. We conduct experiments using a convection based LHTES, under varying operating temperatures, and measure the melted fraction using time-lapse imaging, and first obtain experimentally the dimensional performance curves for a rectangular LHTES. To develop a general-purpose performance curve, we hypothesize the shape of the dimensionless performance curve and the appropriate timescale. A dimensionless performance curve is required to predict device performance at unknown conditions not tested experimentally, and for efficient scale up of the LHTES. We verify our hypothesis experimentally for a convection driven LHTES, with buoyancy driven melting in a rectangular LHTES device. Based on the device geometry and operating Rayleigh number Ra, it is hypothesized that due to convective mixing, the dimensionless time τ taken to reach a discharge state should increase linearly with discharge state, until the amount of solid PCM is small and does not control the nominal temperature difference. A similar behavior is expected for the liquid fraction η, which is analogous to discharge state, but as a measure of latent energy absorbed to total latent energy. The value of melted fraction at which this happens is termed ηcritical, and from numerical results, is observed to be fixed for a given geometry and material with fixed viscosity. This is in contrast to a conduction driven LHTES, where the melting pr