Recently, the ensemble Kalman filter method (EnKF) has been successfully applied in assisted history matching. It is a very powerful tool and can handle a very large number of reservoir parameters, up to millions. The method is a Monte Carlo type formulation, and employs a covariance matrix and maximum a posteriori estimate theory. This new technology has been implemented into an industrial parallel reservoir simulator.

In this paper, we show its application to optimize two horizontal wells equipped with an inflow control valve (ICV) in a sector model of 200,000 cells. One hundred layers were used to capture geological heterogeneity. The two wells were drilled parallel to the edge-water boundary. The optimization objective is to minimize cumulative water production over a 10 year production period.

Encouraging results have been obtained. Cumulative water production was reduced 50% after only five optimization iterations. The optimized ICV opening variation over the entire production history is described. Verified by simulated water production results, the ensemble method did a satisfactory optimization job, better than a trial and error approach could provide. The fully-automated optimization process can be completed within a few hours under a parallel computing environment.

Keywords:
Completion Monitoring Systems/Intelligent Wells, produced water management and control, Artificial Intelligence, history matching, reservoir simulation, optimization problem, objective function, ensemble-based method, well 1, reservoir simulator
Subjects:
Reservoir Simulation, Information Management and Systems, Produced water management and control, History matching, Artificial intelligence, Completion Monitoring Systems/Intelligent Wells, Well Operations and Optimization, Flow control equipment, Well Operations, Optimization and Stimulation
Copyright 2009, Society of Petroleum Engineers
You can access this article if you purchase or spend a download.