期刊
JOURNAL OF ENVIRONMENTAL CHEMICAL ENGINEERING
卷 4, 期 1, 页码 211-218出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.jece.2015.11.008
关键词
Hydrogen sulfide solubility; Gas removal; Gas-liquid equilibrium; Ionic liquids; Artificial neural networks
资金
- National Council for Scientific and Technological Research (CONICYT) [FONDECYT 1150802]
- Direction of Research of the University of Concepcion [DIUC 211.011.054-1.0]
Artificial neural networks have been used for the correlation and prediction of solubility data of hydrogen sulfide in ionic liquids. The solubility of hydrogen sulfide is highly variable for different types of ionic liquids at the same temperature and pressure and its correlation and prediction is of special importance in the removal of hydrogen sulfide from flue gases for which effective and efficient solvents are required. Several network architectures were tested to finally choose a three layer network with 6,10 and 1 neuron, respectively (6, 10, 1). Twelve binary hydrogen sulfide + ionic liquids mixtures were considered in the study. Solubility data (pressure, temperature, gas concentration in the liquid phase) for these systems were taken from the literature (392 data points for training and 104 data points for testing). The training variables are the temperature and the pressure of the binary systems being the target variable the solubility of hydrogen sulfide in the ionic liquid. Average absolute deviations are lower than 4.0% and the maximum individual absolute deviation in solubility is 12.6%. The proposed neural network model is a good alternative method for the estimation of solubility of hydrogen sulfide in ionic liquids for its use in process analysis, process design and process simulation. (C) 2015 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据