A simple and flexible correlation for predicting the viscosity of crude oils

Konstantinos Kotzakoulakis*, Simon C. George

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

10 Citations (Scopus)

Abstract

A simple and flexible correlation has been developed for the prediction of the kinematic viscosity of crude oils based on the Walther equation (1931). The correlation was developed in order to improve on the accuracy of existing correlations and to assist oil spill weathering models, reservoir models and other models where the knowledge of viscosity of uncharacterised crude oils is required. The data used to build the correlation consist of measurements from 137 crude oils from various locations around the world, with 254 viscosity measurements taken at two temperatures, 0 °C and 15 °C, resulting in kinematic viscosity values ranging from 2 cSt to 9,000,000 cSt. The correlation can be used in three different ways depending on the available data. When a single viscosity measurement is available, it can predict the viscosity at a different temperature with an average absolute deviation (AAD) of 13.7% for the studied temperature range. When the specific gravity (SG) and the 50% mass boiling point is available it can give a viscosity estimation with an AAD of 52.9%, which is an improvement in accuracy by a factor of two compared to previously published correlations with the same inputs. Finally, when only the 50% mass boiling point is available, we have improved the accuracy of the Mehrotra (1995) correlation by 10%.

Original languageEnglish
Pages (from-to)416-423
Number of pages8
JournalJournal of Petroleum Science and Engineering
Volume158
DOIs
Publication statusPublished - 1 Sept 2017

Keywords

  • viscosity correlation
  • crude oil
  • oil spill modelling
  • petroleum properties
  • oil weathering
  • Walther equation

Fingerprint

Dive into the research topics of 'A simple and flexible correlation for predicting the viscosity of crude oils'. Together they form a unique fingerprint.

Cite this