An intelligent approach to investigate the effects of container orientation for PCM melting based on an XGBoost regression model
Küçük Resim Yok
Tarih
2024
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Elsevier Sci Ltd
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
The orientation of the container filled with phase change material (PCM) is a critical parameter that significantly effects the performance of thermal energy storage systems. In this study, the Computational Fluid Dynamics (CFD) method is utilised to analyse the effects of container position on the melting process of PCM. Unlike conventional methods, the melting process of PCM was conducted using the hot air jet impingement method. The study investigated the impact of two various Reynolds numbers (2235 and 4470) and three different H/D ratio (the ratio of the distance between the jet and the container to the container diameter) which were 0.4, 0.5, and 0.6, on the PCM melting process. In addition, regression analysis was executed using the Extreme Gradient Boosting algorithm (XGBoost). The outcomes unveiled that the artificial intelligence model attained a minimum accuracy of 97.89 % and reached a maximum accuracy of 99.35 % across the 12 datasets for comparing performance metrics. These results serve as a testament to the prowess of the XGBoost algorithm in providing precise predictions of the target variable within a notably extensive range of accuracy for the datasets under consideration.
Açıklama
Anahtar Kelimeler
Phase change material, Hot air jet impingement, Thermal energy storage, CFD, Container position, XGBoost, Artificial intelligence
Kaynak
Engineering Analysis With Boundary Elements
WoS Q Değeri
N/A
Scopus Q Değeri
Q1
Cilt
161
