Artificial intelligence-driven analysis of dynamic melting in open shell-and-tube latent-heat storage: Effects of PCM inlet pressure, port geometry, and positioning

Abstract

This numerical study explores dynamic melting as an enhancement strategy to improve heat transfer in thermal energy storage (TES) systems utilizing phase change materials (PCM) with openings. Optimizing such systems is crucial for advancing renewable energy storage and integration. A 3D model simulates RT35 PCM flowing through a shell-and-tube heat exchanger annulus. The effects of varying PCM inlet slot diameter (2.5–7.5 mm), inlet pressure (1–40 Pa), and inlet/outlet port positioning on melting fraction and temperature distributions are computationally evaluated. Results show that increasing slot diameter from 2.5 mm to 7.5 mm reduces melting time by 13.6 % (from 550 to 475 min). Raising inlet pressure from 10 Pa to 40 Pa cuts melting time by 43.8 % (from 400 to 225 min). However, varying inlet/outlet port positioning has negligible impact. Larger slot diameters and higher pressures produce more homogeneous melting across the radial width. A neural network approach revealed that port diameters around 5.5 mm maximized melting volume fraction at early stages, while increasing port size accelerated melting in later stages. This study provides quantitative data demonstrating that optimizing dynamic melting parameters through the annulus can substantially enhance heat transfer in open PCM-based TES systems. The novelty lies in systematically evaluating the effects of these design parameters on melting behavior and employing artificial intelligence to further analyze parameter interactions.