Prediction of Octane Number and Additives for Gasoline Blends Using Artificial Neural Networks

فاقد چكيده

Gasoline blending is an important unit operation in gasoline industry. A reliable model for the gasoline blending is beneficial for operation and prediction of gasoline qualities. Since the blending does not follow the ideal mixing rule in practice, Artificial Neural Network (ANN) models have been developed to determine the Research Octane Number (RON) of the gasoline blends produced in Tabriz refinery. The developed ANN models use as input variables the volumetric content of six most commonly used fractions in gasoline productions multiplied by their octane number. In all additives that are used for correcting gasoline octane number, MTBE is the most important component. Economical value ofMTBE in comparison to the other additives, political problems and the government's policy in gasoline production is achieving minimum amount of octane number that specified in the N.I.O.D.C. (National Iranian Oil Refining & Distribution Company) standards. In these standards, 87 is determined as the lower limit of octane number. Simulation results show that ANN models are powerful tools for predicting RON and additives in a specified octane number as judged by R2, MSE, and AARE.