Image-based anisotropic modeling of paraffin melting process for thermal energy storage

Phase change materials (PCMs) are widely used in thermal energy storage (TES) systems due to their high latent heat and ability to maintain nearly constant temperatures during phase change. In particular, the mushy zone plays a crucial role in governing heat and mass transfer during phase change process. However, most existing numerical models oversimplify the mushy zone using isotropic and empirical formulations, often assuming constant value for the mushy zone coefficient Amush. This study presents a morphology-informed modeling framework that couples high-resolution experimental imaging with physical parameter extraction to improve the representation of directional flow resistance in the mushy zone. Using grayscale analysis of high-speed images during the melting of paraffin, direction-specific characteristic lengths and liquid fractions were quantified. These morphological parameters were then used to construct anisotropic permeability models and shape factor corrections. A direction-dependent expression for Amush was formulated and integrated into the enthalpy-porosity method. The results show that the anisotropic model proposed in this study outperforms the isotropic model in terms of the accuracy of predicting the liquid phase fraction (Amush = 107), with the RMSE reduced by 0.0049 and the MAPE reduced by 0.63 %. This morphology-informed approach provides a physically consistent and adaptable representation of flow resistance, enabling more accurate simulation of PCM melting processes.