http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Viscosity estimation of binary mixtures of ionic liquids through a multi-layer perceptron model
Pablo Díaz-Rodríguez,John C. Cancilla,Gemma Matute,José S. Torrecilla 한국공업화학회 2015 Journal of Industrial and Engineering Chemistry Vol.21 No.1
Ionic liquids (ILs) are chemical compounds which are being more and more employed due to theiradvantageous properties which are useful in many chemical and industrial fields. The study of thephysicochemical properties of ILs and their mixtures is essential in order to improve their efficacy. Oneway to study these properties is by designing powerful mathematical estimative models. To do so, anartificial neural network multilayer perceptron (MLP) model has been proposed and developed toestimate the viscosity of four IL binary mixtures (1-hexyl-3-methylimidazolium tetrafluoroborate([C6MIM,BF4]) + 1-ethyl-3-methylimidazolium tetrafluoroborate ([C2MIM,BF4]), [C6MIM,BF4] + 1-butyl-3-methylimidazolium tetrafluoroborate ([C4MIM,BF4]), [C4MIM,BF4] + 1-butyl-3-methylimidazoliummethylsulfate ([C4MIM,MeSO4]), and [C4MIM,BF4] + 1-butyl-3-methylimidazolium hexafluorophosphate([C4MIM,PF6])) inside the temperature range 298.15–308.15 K. The statistical results offered by thedesigned MLP model confirmed that it can be used to accurately estimate the viscosity of the tested ILmixtures (mean prediction errors around 1.5% for a k-fold cross-validation and about 2.3% for three blindtests). Additionally, it is possible to assess the purity level of the studied ILs or binary mixtures throughthe estimated viscosity values.