Integrated software tool brings speed, reliability and greater collaboration to reservoir modeling.
The concept of closed loop reservoir management has been discussed in the literature for decades, yet the traditional approach persists in treating static and dynamic modeling as separate tasks with asset teams continuing to work in domain siloes. This not only increases the time to build a single model but also results in models that might perfectly match the current dynamic data but completely fail to honor the input data and the geological concept. In addition, uncertainty studies are often done in the static domain without being shaped by the information from dynamic data. This increases the risk that key uncertainties are overlooked and reservoir performance is suboptimal.
Using the ResX tool, integrated reservoir modeling workflows were established for the Goliat Field in the Barents Sea offshore Norway. The solution combines data (both static and dynamic), the subsurface know-how of the asset team, reservoir physics, and fit-for-purpose machine learning algorithms to generate a full ensemble of reservoir models that capture subsurface uncertainty. The solution enables frequent model updates when new data arrives, and an automated workflow assists in analyzing results in support of further reservoir management decisions.
Reservoir modeling and data conditioning were carried out in a single step with multidisciplinary engagement forming a streamlined process that could be run on a continuous basis through repeatable workflows.
An efficient model updating workflow facilitated a systematic analysis of the ensemble and the evaluation of lookahead development options (e.g. where to drill infill wells, what changes to operational strategies, etc.). This enabled the exploration of a variety of reservoir management strategies to maximize the field’s potential.
An ensemble-based modeling approach at the Goliat Field demonstrated how computer power, streamlined data integration, and subsurface team expertise can be brought together to improve reservoir understanding in a fraction of the time needed with traditional tools and approaches. Among salient features, it was possible to automatically create infill wells within the identified targets, taking into account physical constraints such as dogleg severity.
References: Panfili, P., et al., 2017. Integrated Software Tool Brings Speed, Reliability to Reservoir Modeling on Barents Sea Project. Journal of Petroleum Technology
Reduce well construction costs with the latest-generation of filtration services, wellbore cleanup fluids, mechanical wellbore cleaning tools, and optimized software modeling.
