Scholarlink

<p>The primary objective of employing phase change material (PCM) is to improve thermal inertia and reduce energy consumption for heating and cooling applications. Considering the real field applications of PCMs in construction, the current investigation presents experimental and predictive modelling of the thermal performance of palmitic acid (PA) encapsulation in a silica shell integrated into waste-derived alkalinated fly ash ground granulated blast-furnace slag (GGBS) geopolymer PA@SiO<sub>2</sub> (GP<sub>2</sub>) and subsequently measured the thermal profile and compared to neat PCM (GP<sub>1</sub>). The thermal stability of PA@SiO<sub>2</sub> effectively lowered the inner temperature to an average of 8&#xa0;°C by overcoming leakage of PA and increased thermal absorption in contrast to that of PA. The microporosity of PA and PA@SiO<sub>2</sub> composite was effectively alleviated through the binding interface of graphite, resulting in improved thermal conductivity. The high enthalpy of PA@SiO<sub>2</sub> (130 Jg<sup>−1</sup>), with an encapsulation efficiency of 80%, led to an improved system efficiency. Decision tree, random forest, XGBoost, LightGBM, CatBoost, SVR, KNN and ANN machine learning models were employed to compare the different performance metrics determined. According to evaluation metrics, ANN model outperformed the other models. Thus, leveraging the predicted thermal performance results helps to automatically identify the optimal matrix endowed with the combined capabilities of thermal energy storage and thermal conductivity. This study emphasizes the advancement of predictive machine learning models to analyse the effective temperature regulation performance of graphite-embedded GP<sub>2</sub> matrix.</p> Graphical abstract <p></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

Machine learning-based analysis of thermal performance in graphite-integrated fly ash-GGBS geopolymer with palmitic acid phase change material

  • 期刊论文

  • 发表日期：

    2026年4月25日

查看全文
  • Rithikaa Thanigaiselvan,
  • B. Kalidasan,
  • R. Jeyalakshmi

摘要

The primary objective of employing phase change material (PCM) is to improve thermal inertia and reduce energy consumption for heating and cooling applications. Considering the real field applications of PCMs in construction, the current investigation presents experimental and predictive modelling of the thermal performance of palmitic acid (PA) encapsulation in a silica shell integrated into waste-derived alkalinated fly ash ground granulated blast-furnace slag (GGBS) geopolymer PA@SiO2 (GP2) and subsequently measured the thermal profile and compared to neat PCM (GP1). The thermal stability of PA@SiO2 effectively lowered the inner temperature to an average of 8 °C by overcoming leakage of PA and increased thermal absorption in contrast to that of PA. The microporosity of PA and PA@SiO2 composite was effectively alleviated through the binding interface of graphite, resulting in improved thermal conductivity. The high enthalpy of PA@SiO2 (130 Jg−1), with an encapsulation efficiency of 80%, led to an improved system efficiency. Decision tree, random forest, XGBoost, LightGBM, CatBoost, SVR, KNN and ANN machine learning models were employed to compare the different performance metrics determined. According to evaluation metrics, ANN model outperformed the other models. Thus, leveraging the predicted thermal performance results helps to automatically identify the optimal matrix endowed with the combined capabilities of thermal energy storage and thermal conductivity. This study emphasizes the advancement of predictive machine learning models to analyse the effective temperature regulation performance of graphite-embedded GP2 matrix.

Graphical abstract