Temizel, Cenk (Aera Energy) | Zhiyenkulov, Murat (Schlumberger) | Ussenova, Kamshat (Schlumberger) | Kazhym, Tilek (Embamunaygas) | Canbaz, Celal Hakan (Schlumberger) | Saputelli, Luigi Alfonso (Frontender Corporation)
Optimum well placement in intelligent fields, using previously developed optimal control methods to maximize net present value (NPV), is becoming practical with recent advances in technologies as well as their applications to the petroleum industry. To efficiently use these methods in an intelligent field, an assessment of its economic aspects and its performance, especially in reservoirs with high degree of heterogeneity (uncertainty), must be made. By using such integrated workflows, mature and new field can be developed better. The workflow could be used as a reliable tool for improving the decision-making process.
There are multiple optimization techniques used in the industry for optimizing well placement (e.g. direct and gradient optimization). With the use of reservoir simulation case study, this paper aims to provide a comparative performance analysis of multiple optimization techniques. To make the evaluation stronger and more application to a real-world problem, the model selected for this study has a high degree of geological uncertainty and constraints for computation time, infrastructure and complexity to decide on optimal well placements.
Having a better understanding on the uncertainties in geology lead to more robust decisions in reservoir management. Right strategy especially helps in optimizing larger scale, million-cell model simulations enabling practical implementation of reservoir simulation coupled with optimization.
Optimum well placement in complex reservoirs requires a complete grasp of optimization methods, key factors and constraints but most importantly the effect of geological uncertainty. A lack of awareness of optimization algorithms and their applications by engineers is a drawback in this process. In addition, complete evaluation of geological uncertainty is another challenge. This study provides an understanding and clarification to serve as a guideline on optimization practices by outlining the significant components in the process.