Sheikhshoaei Amir Hossein, Sanati Ali
Petroleum and Petrochemical Engineering School, Hakim Sabzevari University, Sabzevar, Iran.
Sci Rep. 2025 Jul 2;15(1):22672. doi: 10.1038/s41598-025-08947-7.
Ionic liquids (ILs) as eco-friendly solvents have gained significant attention across various fields of science, including the petroleum industry. Among the different ILs families, imidazolium-based ILs have been the subject of many research studies. However, not enough experimental studies have been conducted to determine the viscosity of this family of ILs, making accurate viscosity prediction crucial for their practical applications. This study aims to predict the viscosity of imidazolium-based ILs and their mixtures using critical properties as input parameters. Machine learning (ML) models have been implemented, and their performance in viscosity prediction for IL mixtures was compared with a molecular-based model, ePC-SAFT-FVT (ePC-FVT-MB), and an ion-based model, ePC-SAFT-FVT (ePC-FVT-MB). Graphical and statistical analyses revealed that the RF model offered the lowest error for viscosity prediction of pure ILs, while CatBoost performed the best for IL mixtures. In addition, sensitivity analysis showed that viscosity decreased with temperature and increased with pressure. The proposed models exhibit high accuracy under varying conditions. Outlier detection using the Leverage method indicated that 95.11% of pure IL viscosity data and 94.92% of mixed ILs viscosity data are statistically valid.
离子液体（ILs）作为环境友好型溶剂，在包括石油工业在内的各个科学领域都受到了广泛关注。在不同的离子液体家族中，咪唑基离子液体一直是许多研究的主题。然而，尚未进行足够的实验研究来确定该家族离子液体的粘度，因此准确预测粘度对于其实际应用至关重要。本研究旨在以临界性质作为输入参数，预测咪唑基离子液体及其混合物的粘度。已实施机器学习（ML）模型，并将其在离子液体混合物粘度预测中的性能与基于分子的模型ePC-SAFT-FVT（ePC-FVT-MB）和基于离子的模型ePC-SAFT-FVT（ePC-FVT-MB）进行了比较。图形和统计分析表明，随机森林（RF）模型在纯离子液体粘度预测中误差最低，而CatBoost在离子液体混合物预测中表现最佳。此外，敏感性分析表明，粘度随温度降低而升高，随压力升高而降低。所提出的模型在不同条件下均表现出高精度。使用杠杆法进行的异常值检测表明，95.11%的纯离子液体粘度数据和94.92%的混合离子液体粘度数据在统计上是有效的。
