Reliable estimate of minimum miscibility pressure from multiple possible EOS models for a reservoir oil under data constraint

Abstract

Multiple Equation of State (EOS) models are possible for a reservoir oil if limited laboratory data is used in the EOS model development; each EOS model has its own predicted minimum miscibility pressure (MMP) value with unknown reliability. This article presents a novel approach to estimate a reliable MMP value using the multiple MMP values calculated from multiple EOS models developed with limited laboratory data such as bubble point and density at bubble point. MMP predicted using the proposed approach has better accuracy than the estimated MMP from an EOS model developed using a large set of data at the regression stage. Each possible EOS model for a reservoir oil has a specific set of values of total Gibbs free energy of mixing (g) at bubble point and predicted MMP for a given injection gas. The MMP estimate is developed using the trend of g versus MMP (predicted) from multiple EOS models in the proposed approach. The AARD in MMP prediction for 22 oils is 5.21% using SRK EOS and 6.65% using PR EOS, which is lower than the earlier approach (7.6%) that uses all PVT, swelling test, and multi-contact test data in EOS model development.

Keywords:

• EOS model
• Gibbs free energy
• minimum miscibility pressure estimation
• miscible gas injection
• reservoir fluid characterization

Acknowledgments

The authors acknowledge the financial assistance provided by the Indian Institute of Technology (Indian School of Mines), Dhanbad, for this research (FRS(107)/2016-2017/PE).