A correlation to predict the viscosity of light crude oils

Abstract

Direct viscosity measurements are often expensive or unavailable. Therefore, empirical correlations are often used for predicting the viscosity of crude oils. However, several published correlations are either too simplistic or too complex for routine operational use. Many of the common correlations in use were developed using data from other regions of the world, Empirical correlations for predicting the viscosity of light crude oils in the Niger Delta have been presented in this paper. Data from over 400 oil reservoirs from the Niger Delta were collected. The samples were representative of the two crude oil viscosity regimes: above and below the bubble point. After normal quality checks, non-linear multiple regression with linear partial correlation coefficient techniques were used to establish simple correlations between viscosity, pressure, temperature, oil specific gravity and solution gas oil ratio. Statistical error analysis of the developed correlation showed average absolute relative percentage error of 4.00% and 3.25% and R2 of 0.99 and 0.97 for oil viscosity above and below the bubble point respectively. These results constitute considerable improvements over existing correlations.