Transient thermal network model for rapid multi-objective optimization of plate-fin latent heat thermal energy storage unit

Latent heat thermal energy storage represents a promising solution for thermal management of electronic devices. However, optimal design of latent heat thermal energy storage units (LHTESUs) is often hindered by high computational costs associated with transient numerical simulations. In this work, we propose a threedimensional transient thermal network model (TNM) that simplifies the complex transient heat transfer equations into a system of linear equations. TNM significantly reduces computational time by more than 99.9 % compared to the finite volume method (FVM). During the melting process, the maximum mean absolute error for the liquid fraction and outlet temperature are found to be 0.033 and 1.45 degrees C, respectively, indicating minimal deviation. A multi-objective optimization is performed by integrating TNM with genetic algorithms. The optimized LHTESU demonstrates a 23.1 % increase in average effectiveness, a 53.9 % increase in energy storage density, and a 25.5 % increase in specific power. This study opens a new route for rapid simulation and optimization of LHTESUs.