IoT enabled deep learning and multi objective optimization of nanoPCM solar thermal energy storage for intelligent charging discharging control

Abstract

Temporal differences between the availability of solar thermal energy and thermal requirements limit the effective utilization of solar thermal power, necessitating advanced thermal energy storage solutions. This study complemented the working performance of phase change material (PCM) based thermal energy storage system with the assistance of graphene nanoparticle and intelligent optimization approach.An experimentally studied PCM storage unit based on paraffin was embedded with a helical heat exchanger and tested under real outdoor conditions with graphene nanoparticle loadings of 0, 0.1, 0.3, and 0.5 wt%. The methodology is the combination of the experimental study and the Artificial Neural Network (ANN) surrogate modelling and multi-objective optimization with the assistance of NSGA-II in order to find Pareto-optimal operating conditions.Experimental evidence shows that the addition of graphene increases the charging rate of PCM by 30–35%, enhances stored energy by 10–18%, and extends useful discharge time by 25–30% compared to pure PCM. Improved performance of the energy storage is demonstrated by a drastic increase in cycle thermal efficiency (up to approximately 74%) and exergy efficiency and both quantitative and qualitative. The novelty of the work is that it integrates Nano-enhanced PCM and AI-controlled prediction and optimization of adaptive solar thermal energy systems.The combined forecasting–optimization approach improves discharge stability and energy recovery under varying solar conditions, highlighting the potential of smart Nano-PCM storage systems.