The OU Consortium on Ensemble Methods (OU -- CEM) conducts research and development on the use of ensemble methods for history matching of reservoir simulation models, continuous model updating, and optimal control of reservoir processes.

Development of efficient and robust algorithms based on ensemble Kalman filter (EnKF) for reservoir data assimilation, automatic history matching, and uncertainty quantification; Development of techniques to make EnKF robust for small ensemble size or with approximate models; Investigation of the efficiency and accuracy of the EnKF vs. other methods for automatic history matching, data assimilation and parameter estimation, in the context of multi-phase, compositional, and black oil reservoir models under realistic reservoir conditions; Exploration of innovative filters such as Karhunen-Loeve expansion based Kalman filter, resampling and particle filters for improving the performance of EnKF under strong non-linearity and large dimensionality; Development of ensemble methods for reservoir management optimization; Demonstration of developed methodologies and algorithms on real world reservoir production problems.

Research within the group is directed by Dr. Dean S. Oliver, Mewbourne Chair Professor in the Mewbourne School of Petroleum and Geological Engineering, OU.