Title

An accurate model to predict viscosity of heavy oils

Document Type

Article

Publication details

Jahani-Keleshteri, Z & Bahadori, A 2017, 'An accurate model to predict viscosity of heavy oils', Petroleum Science and Technology, vol. 35, no. 4, pp. 371-376.

Published version available from:

http://dx.doi.org/10.1080/10916466.2016.1259632

Peer Reviewed

Peer-Reviewed

Abstract

The viscosity of heavy oils is a crucial factor in estimating oil recovery. Viscosity plays a vital character in reservoir simulations as well as in estimating the easiness of fluid flow, estimating oil recovery, and choosing a production model. The authors present a new intelligent model, a GA-LSSVM, used for predicting the viscosity of heavy oils. The experimental data employed in this work are the product of searching in many heavy oils data gathered from the literature. Development of robust predictive models to predict the viscosity of heavy oils is of immense help in many process engineering applications. The outcomes revealed that GA-LSSVM is able to capture the complicated and nonlinear relation between the input and output variables. GA-LSSVM model resulted in R2 and mean absolute error values of 0.9999 and 0.9924, respectively.