Experimental Analysis and Neural Network Model of MWCNTs Enhanced Phase Change Materials

In this article, the thermophysical properties of binary eutectic PCM salt (LiNO 3 + NaCl) are investigated experimentally using the dispersion of multi-walled carbon nanotubes (MWCNTs) with varying weight fractions (i.e., 0.25 %, 0.5 %, and 1 %). According to the XRD and FTIR results, MWCNTs physically amalgamated with base eutectic salt without affecting their chemical structure. The Kissinger model is used to assess the phase change kinetics of the prepared nano-PCM composites. With the addition of MWCNTs, the activation energy of the chosen PCM is significantly increased. The thermophysical properties of the nano-PCM samples, such as phase transition temperature and latent heat value, are measured using differential scanning calorimetry (DSC) and thermal conductivity using a laser flash apparatus. The results show that dispersing 1 % MWCNTs in PCM salt improved the thermal conductivity enhancement ratio by 38.59 %, while decreasing the latent heat storage capacity by 14.98 %. Furthermore, by training the experimental DSC values of nano-PCM samples at various heating rates, an artificial neural network is developed to predict the thermophysical properties of nanocomposites. With an R 2 value of 0.998, the developed neural network accurately predicted the experimental DSC values.