Saman Jahanbakhshi - Chemical & Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran
Mahmoud Reza Pishvaie - Chemical & Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran
Ramin Bozorgmehry Boozarjomehry - Chemical & Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran

Accurate description of multiphase flow in porous media is crucial for practical applications like enhanced oil recovery and environmental remediation of the unsaturated zone. Multiphase flow models are conventionally based on the extension of Darcy’s law and the only way to put the real physics of the flow into these conventional models is through relative permeabilities. This study considers application of the ensemble-based assimilation methods, specifically ensemble Kalman filter (EnKF) and evolutionary data assimilation (EDA), to estimate relative permeability curves in the oil-water and the gas-oil systems from measured historical data. Accurate estimation of relative permeabilities is obtained by assimilating noisy production rates and bottomhole pressures. The EnKF is a recursive filter and is most suitable for problems with a large number of variables because it is easy to implement and computationally efficient. However, in the EDA, a smaller population evolves over several generations and this improves parameter estimation convergence.

Three-phase flow, Relative permeability, Ensemble Kalman filter, Evolutionary data assimilation