شماره ركورد كنفرانس :
3834
عنوان مقاله :
An artificial neural network model for the viscosity prediction of binary mixtures of ionic liquid 1-butyl-3-merthylimidazolium tetrafluoroburate and polyethylene glycol with different average molecular weights
پديدآورندگان :
Daneshvar Azadeh a.daneshvar88@yahoo.com Faculty of Chemistry, Department of Physical Chemistry, University of Isfahan, Isfahan, Iran; , Moosavi Majid Faculty of Chemistry, Department of Physical Chemistry, University of Isfahan, Isfahan, Iran
كليدواژه :
Artificial neural network , Ionic liquid , Polyethylene glycol , Viscosity
عنوان كنفرانس :
نوزدهمين سمينار شيمي فيزيك ايران
چكيده فارسي :
In this work, the viscosity of four different binary mixtures of 1-butyl-3-methylimidazolium tetrafuoroburate and poly ethylene glycol with average molecular weights of 200, 400, 600 and 1000 have been measured at different compositions, temperatures and shear rates. An artificial neural network method has been developed to model simultaneously the shear-, temperature-, composition- and average molecular weight of poly ethylene glycol dependencies of the viscosities in the studied mixtures. Total number of experimental data used to design the stated network were 615 data points. The results show that the optimal number of neurons in the hidden layer was 24 and the best network topology was obtained as (4-24-1).
