Viscosity Reduction of Heavy Crude Oil by Dilution Methods: New Correlations for the Prediction of the Kinematic Viscosity of Blends

• Main Authors: Saeed Mohammadi, Mohammad Amin Sobati, Mohammad Sadeghi
• Format: Article
• Language: English
• Published: Petroleum University of Technology 2019-01-01
• Series: Iranian Journal of Oil & Gas Science and Technology
• Subjects: heavy crude oil; kinematic viscosity; blending; genetic algorithm; binary blend
• Online Access: http://ijogst.put.ac.ir/article_55719_47d1c328365b2d960c0e21ccafe21eaf.pdf

Dilution is one of the various existing methods in reducing heavy crude oil viscosity. In this method, heavy crude oil is mixed with a solvent or lighter oil in order to achieve a certain viscosity. Thus, precise mixing rules are needed to estimate the viscosity of blend. In this work, new empirical models are developed for the calculation of the kinematic viscosity of crude oil and diluent blends. Genetic algorithm (GA) is utilized to determine the parameters of the proposed models. 850 data points on the viscosity of blends (i.e. 717 weight fraction-based data and 133 volume fraction-based data) were obtained from the literature. The prediction result for the volume fraction-based model in terms of the absolute average relative deviation (AARD (%)) was 8.73. The AARD values of the binary and ternary blends of the weight fraction-based model (AARD %) were 7.30 and 10.15 respectively. The proposed correlations were compared with other available correlations in the literature such as Koval, Chevron, Parkash, Maxwell, Wallace and Henry, and Cragoe. The comparison results confirm the better prediction accuracy of the newly proposed correlations.