Decision Based History Matching: A Quick, Efficient, and Effective Way to Explain and Predict Unexpected Water Breakthrough and Trends in the W10 Reservoir, Dotun Field, Deepwater Nigeria

Abstract

Abstract History matching is an inverse modeling technique in which production-driven changes in reservoirs (pressure/saturation) are used to calibrate reservoir model properties and thus narrow down the associated uncertainty ranges, such that historical dynamic data is honored within acceptable tolerance, which results in subsurface models that are adequate for production forecasts & development planning. This paper summarizes an iterative history matching technique carried out to better understand early water breakthrough and observed water-cut trend in the W10 reservoir of Dotun field developed through a recently drilled producer-injector well-pair. Early water breakthrough & high water-cut seen in the producer well can be explained by the existence of a strong aquifer and a high permeable conduit between the producer and the aquifer which may not be resolved by seismic. This assumption is supported by interference tests showing weak pressure communication between producer/injector and chemical analysis indicating that the formation is the source of the produced water. A decision based modeling approach was adopted whereby facies updates were iteratively carried out in the sector between the producer and the aquifer to achieve incremental improvement in Water Breakthrough Time and Water-Cut matches. Dynamic parameters like possible Contact uncertainty, Relative Permeability and Skin were also varied. For accurate flood front modelling, local grid refinement techniques were implemented. A reasonable history match was achieved with a shallower contact, a facies realization replicating continuation of a high permeability streak between producer and aquifer, significant aquifer strength and dynamic local grid refinement. The model was history matched with four months of post-water-breakthrough production data and the robustness of the technique has been validated by actual well performance (post study) with simulation predictions. The exercise also suggests that the pressure support at current production rates is maintained mostly by the aquifer and that the injector could indeed be redundant.