Scholarlink

<p>Deep Eutectic Solvents (DESs) have garnered significant interest due to their diverse applications; however, detailed thermophysical data, especially for density and viscosity, are needed to expand their industrial use. This study developed an Extreme Gradient Boosting (XGBoost) model to predict the density and viscosity of DESs. Experimental measurements were conducted on 494 density data points (308–353&#xa0;K) and 1600 viscosity data points (293.15–323.15&#xa0;K) from 40 DESs synthesized by varying the molar ratios of three hydrogen bond acceptors and donors. The dataset was used to train an Extreme Gradient Boosting model (XGBoost) for accurate property prediction. The XGBoost model demonstrated high accuracy, with a mean absolute percentage error (MAPE) of 0.16% for density and 1.5% for viscosity. These results indicate that the model is a reliable tool for predicting key thermophysical properties, facilitating the broader application of DESs in various industries.</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

High-Precision Estimation of DES Density and Viscosity using Extreme Gradient Boosting: Experimental Insights and Modelling

  • 期刊论文

  • 发表日期：

    2025年10月6日

查看全文
  • Krunal J. Suthar,
  • Atir Sakhrelia,
  • Amaan Mansuri,
  • Anaya Patel,
  • Priyank Thakkar,
  • Milind H. Joshipura

摘要

Deep Eutectic Solvents (DESs) have garnered significant interest due to their diverse applications; however, detailed thermophysical data, especially for density and viscosity, are needed to expand their industrial use. This study developed an Extreme Gradient Boosting (XGBoost) model to predict the density and viscosity of DESs. Experimental measurements were conducted on 494 density data points (308–353 K) and 1600 viscosity data points (293.15–323.15 K) from 40 DESs synthesized by varying the molar ratios of three hydrogen bond acceptors and donors. The dataset was used to train an Extreme Gradient Boosting model (XGBoost) for accurate property prediction. The XGBoost model demonstrated high accuracy, with a mean absolute percentage error (MAPE) of 0.16% for density and 1.5% for viscosity. These results indicate that the model is a reliable tool for predicting key thermophysical properties, facilitating the broader application of DESs in various industries.