Scholarlink

<p>This study investigated the melting and solidification of paraffin wax within hexagonal latent heat energy storage units. Each unit features a petal-shaped tube with Y-shaped, non-uniform fins, through which heat transfer fluid flows to charge and discharge the modules. Natural convection effects in the molten phase were accounted for using continuity and momentum equations, with the finite element method simulating the phase change. A dataset of 145,973 simulated cases was then used to train an Artificial Neural Network (ANN) to learn the system’s phase change behavior. This trained ANN subsequently generated control parameter maps for both melting and solidification. The results show that shifting the petal and fins downward enhances melting due to improved natural convection flows, while having a negligible impact on solidification. Fin geometry also plays a crucial role for a constant fin mass. For instance, a design with long lateral fins exhibited 6% longer melting and 16.2% longer cooling times compared to a design with short lateral fins. An optimal Y-angle exists for the middle fin during melting, whereas larger angles prove more beneficial for solidification through enhanced conduction. Increasing the stem fin ratio (reducing Y part) of the middle fin can improve solidification time by 25.5%.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence-numerical simulation of melting and solidification heat transfer in a bundle of hexagonal enclosures enhanced by non-uniform Y-shape fins and petal tubes

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月23日

查看全文
  • Mohammed Aldawsari,
  • Jana Shafi,
  • Pouyan Talebizadehsardari,
  • Mohammad Ghalambaz

摘要

This study investigated the melting and solidification of paraffin wax within hexagonal latent heat energy storage units. Each unit features a petal-shaped tube with Y-shaped, non-uniform fins, through which heat transfer fluid flows to charge and discharge the modules. Natural convection effects in the molten phase were accounted for using continuity and momentum equations, with the finite element method simulating the phase change. A dataset of 145,973 simulated cases was then used to train an Artificial Neural Network (ANN) to learn the system’s phase change behavior. This trained ANN subsequently generated control parameter maps for both melting and solidification. The results show that shifting the petal and fins downward enhances melting due to improved natural convection flows, while having a negligible impact on solidification. Fin geometry also plays a crucial role for a constant fin mass. For instance, a design with long lateral fins exhibited 6% longer melting and 16.2% longer cooling times compared to a design with short lateral fins. An optimal Y-angle exists for the middle fin during melting, whereas larger angles prove more beneficial for solidification through enhanced conduction. Increasing the stem fin ratio (reducing Y part) of the middle fin can improve solidification time by 25.5%.