Accurate prediction of the viscosity of light crude oils using one-parameter friction theory: Effect of crude oil characterization methods and property correlations. (1st January 2021)

• Record Type: Journal Article
• Authors: Khemka, Yash; Abutaqiya, Mohammed I.L.; Sisco, Caleb J.; Chapman, Walter G.; Vargas, Francisco M.
• Is Part Of: Fuel. Volume 283(2021)
• Publication Date: 2021-01-01
• Subjects: Friction theory -- Live oil viscosity -- Crude oil characterization -- Critical property estimation; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2020.118926
• Languages: English
• ISSNs: 0016-2361

Abstract

Highlights: Six friction theory-based models are developed to predict crude oil viscosity. Three characterization methods and two critical property correlations were analyzed. All six models successfully predicted viscosity with an average error below 10%. Choice of characterization method affects the accuracy of viscosity predictions. Abstract: The one-parameter friction theory framework using the Peng-Robinson equation of state (PR FT) (Quiñones-Cisneros et al., Fluid Phase Equilibria, 2001) is applied for the viscosity modeling of light, crude oils. Three different methods have been implemented to characterize and determine the composition of these fluids: SARA-based method using the Perturbed-Chain Statistical Association Fluid Theory (PC-SAFT) EoS (Punnapala and Vargas, Fuel, 2013), SARA-based method using the PR EoS (Abutaqiya et al., Energy & Fuels, 2020), and Single Carbon Number (SCN) method using the PR EoS (Pedersen and Christensen, Taylor & Francis Group, 2007). Both SARA-based methods use the Saturates-Aromatics-Resins-Asphaltenes (SARA) content analysis. Additionally, two different property correlation sets have been used with each characterization method to estimate the critical properties of the generated pseudo-components: Evangelista and Vargas (EV) correlations (Evangelista and Vargas, Fluid Phase Equilibria, 2018) and Pedersen correlations (Pedersen and Christensen, Taylor & Francis Group, 2007). The predictive capabilities of the different modeling schemes are tested against experimental viscosity data for 10 light crude oils from the Middle East after fitting the friction theory parameters to a single viscosity data point at saturation condition. A systematic comparison of the characterization methods revealed that the SARA-based methods with either EoS predict viscosity with higher accuracy (below 5% AAPD) than the SCN method (above 5% AAPD), irrespective of the correlations used. Despite using the relatively simpler PR EoS with SARA-based method, the viscosity predictions are as good as the predictions obtained using the highly advanced PC-SAFT EoS.