The Brage field was discovered in 1980 and started production from its Fensfjord reservoir in 1993.

The Fensfjord Fm. is a highly heterogeneous reservoir with low STOIIP density. Lack of zonal control and several drainage strategies throughout the years make production allocation and drainage history very complex. Thus, achieving a full field history match in the Fensfjord reservoir through deterministic modeling has always been a challenge.

Current STOIIP estimates indicate that there is room for several more producers. However, identification of target areas for future wells are associated with high uncertainties. Using advanced stochastic modeling and ensemble-based methods has reduced the uncertainties and thereby improved the history match quality and understanding of the reservoir.

Bayesian statistical methods have been applied to combine the structural and well path TVD uncertainties, and generate multiple realizations of the top reservoir surface, isochores and well trajectories, serving as additional history matching parameters. Based on this input, an ensemble of static models was iteratively conditioned to the dynamic data resulting in a significant improvement of the history match quality.

Sensitivity analysis have been performed to identify and breakdown the impact of the uncertain parameters on the in-place volumes.

This paper shows how the new integrated approach has increased the predictive power of the model, resulting in identification of several infill targets for future drilling campaigns.

You can access this article if you purchase or spend a download.