科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Applied Thermal Engineering2026-04-20· Materials science

Artificial intelligence-GFEM approach for predicting VIFs impact on HNEPCM solidification in hexagonal latent heat thermal energy storage systems with solar radiation

Mehdi Mahboobtosi, Fateme Nadalinia Chari, Hesam Ehsani, Davood Domiri Ganji

原始摘要(英文原文)· Original abstract
This study explores the impact of Vascular-Inspired Fins (VIFs) with varying configurations on the solidification of Phase Change Material (PCM) in Latent Heat Thermal Energy Storage Systems (LHTESS). The novelty of this work lies in the combined investigation of vascular-inspired fin geometries, penta-hybrid nanoparticle-enhanced PCM, and solar radiation effects, together with the development of an ANN-based surrogate model for rapid and accurate prediction of LHTESS performance. Nine different cases, combining 2, 3, and 4 main fins with 2, 3, and 4 secondary fins, were analyzed, along with the integration of penta-hybrid nanofluid (PHNF) and solar radiation to enhance solidification. An Artificial Neural Network (ANN) was employed as a surrogate model to predict key metrics such as solid fraction, average temperature, and total energy storage. ANN provides a computationally efficient surrogate model with minimal loss of accuracy compared to the full GFEM simulations The results revealed that increasing the radiation parameter (Rd) from 0 to 1 caused a 45.81% reduction in Full Solidification Time (FST), alongside a 1.96% decrease in average temperature and a 1.95% reduction in total energy over 12,000 s. Switching from hybrid nanofluid to ternary hybrid nanofluid led to a 24.26% decrease in FST, 0.9% reduction in average temperature, and 10.33% decrease in total energy. The use of PHNF instead of ternary hybrid nanofluid resulted in a 32.29% reduction in FST, 1.11% decrease in average temperature, and 10.46% reduction in total energy. The ANN model achieved high accuracy, with minimal prediction errors, validating its potential for real-time performance predictions and intelligent control in LHTESS. These findings highlight the effectiveness of combining optimized fin designs, advanced nanofluids, and solar radiation to enhance solidification, offering a promising approach to improving thermal energy storage system efficiency.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Artificial intelligence-GFEM approach for predicting VIFs impact on HNEPCM solidification in hexagonal latent heat thermal energy storage systems with solar radiation — 科研速览 Science Skim