Hydrocarbons density estimates for a wide range of conditions using RBF-ANN and ANFIS strategies

Abstract

ABSTRACTThe volumetric properties of hydrocarbons such as density have particular effects on designing industrial processes so in the current study, the densities of hydrocarbon mixtures have been predicted by utilizing a radial basis function artificial neural network (RBF-ANN) and an adaptive neuro-fuzzy inference system (ANFIS). A number of 2891 actual data of hydrocarbon density in different pressures and temperatures for aromatic, branched, long-chain and short-chain hydrocarbons. An extensive comparison between proposed models outputs and experimental density data has been carried out in different graphical and statistical manners. This work illustrates that RBF-ANN and ANFIS algorithms represent high ability alternatives for prediction of density of hydrocarbons in high pressure and temperature conditions.