Abstract
An integrated workflow was developed to support the waterflood design of an onshore field in Brazil. This giant mature field has more than 2000 drilled wells with a long production history that has been declining. The objective of the study was then to improve the recovery factor for that field, as well as generate an integrated workflow that could be adapted and applied to other similar fields.
The workflow comprised four main stages. It started with the gathering and treatment of all relevant input data, such as fluid and rock lab data, well logs, and production historical data, to construct a simulation model fit for streamline simulation. A sensitivity study was then conducted analysing the uncertain parameters that had most impact on the simulation results, followed by an uncertainty analysis. Best candidates from this second phase were then used as base cases for the history match process. Eventually, the waterflood design was analysed and optimized considering three main aspects: water allocation, workovers and well placement.
The water allocation was first optimized and a reduction of about a fifth of injected water was achieved while maintaining the level of oil production. This was performed using the Pattern Flood Management algorithm (PFM), available in the streamline simulator. This module performed water re-allocation based on bundle efficiency ranking. Different control criteria and optimization parameters were experimented to reach an optimal result. The potential for workovers and, in particular conversion of producers into injectors, was then evaluated but didn't provide a significant improvement in results. Eventually it was considered an increase in well count, looking into optimized well placement based on sweet spot maps and streamline analysis. These solutions were finally combined in an iterative process to ensure interactive effects were accounted for and all aspects jointly optimized and led to an expected increase in oil production of about 5%.
This study generated an integrated workflow bridging a long production history with a full-field simulation model for this large mature field. Also, using streamline simulation for such waterflood design optimization appeared fit for purpose. First, it brought an improved efficiency as the workflow required running several scenarios. Second, it allowed to not only consider traditional tools to improve recovery factor but also solutions making use of the understanding of model connectivity the streamline simulator provides.
