Model-based production strategy optimization for a heavy oil reservoir considering waterflooding and intelligent wells

Heavy oil reservoir production is complex due to low oil recovery factors and the difficulty to flow from the reservoir to the production field system. Decision-making procedures for developing and managing a production strategy are also hard because all variables, uncertainties, and physical phenomena must be studied to avoid potential wrong decisions. Waterflooding (WF) is the most common method to recover oil, but the management in heavy oil reservoirs is difficult due to low sweep efficiency caused by a high water-oil mobility ratio and highwater production. The WF management can be improved by using intelligent wells (IW) equipped with inflow control valves (ICVs) because they can control multiple production/injection zones. The optimization of WF with ICVs as production strategies is also challenging. It requires considerable effort due to the variables and principles to be studied. As for IW with ICVs, extra work is necessary since more design and operational parameters play a significant role in the reservoir management. The objective of this work is to perform a nominal production optimization for the development and management of a heavy oil reservoir considering waterflooding without ICVs (WF) and with ICVs (WF + ICV) as production strategies. A complete methodology is applied to select and compare the strategies by optimizing the design and control variables through model-based reservoir simulation, using the Net Present Value (NPV) as the objective function (OF). We use manual and assisted processes to maximize the OF based on reservoir engineering knowledge and applying the Iterative Discrete Latin Hypercube sampling algorithm (IDLHC). The study case is named EPIC001, which has a 13 API heavy oil reservoir and is part of a Brazilian offshore field. The results showed that WF + ICV is more feasible for our case and obtained a larger NPV. The WF + ICV strategy had a better sweep efficiency than WF due to intelligent management in the completed well intervals provided by the ICV controls. More oil with less production and injection of water, while maintaining the reservoir pressure overcame the lower field performance under WF without the ICVs. Our methodology worked adequately to optimize WF and WF + ICV for a heavy oil reservoir considering a nominal case. Consequently, a decision-maker or a researcher could use this procedure for similar cases to optimize, compare, and select production strategies.