(222a) Dynamic Oil and Gas Production Systems Optimization Via Explicit Reservoir Multiphase Flow Simulation

In an era of globalized business operations, large and small oil and gas producers alike strive to foster profitability by improving the agility of exploration endeavors and the efficiency of crude oil production, storage and transport processes: they face acute challenges, aggressive financial goals and strict environmental constraints, all necessitating a high level of oilfield modeling accuracy, so as to maximize recovery.

Dynamic oil and gas production systems simulation and optimization is a research trend with a potential to meet the foregoing challenges of the oil and gas industry [1]. Production optimization challenges are addressed via Peaceman correlations [2,3] for two-phase flow of oil and gas in production circuits, with a series of assumptions: the cornerstone of the model is the decoupling assumption for the reservoir model, based on the vast timescale differences among the oil production circuit elements. While the surface and pipeline facilities are in principle no different from those of any petrochemical plant, the sub-surface elements (reservoirs, wells) induce complexity that must be studied via a systematic strategy that has not been hitherto proposed.

The complex two-phase flow in production wells governs crude oil and gas transport. Despite intensive experimentation and extensive CFD simulations towards improved understanding of flow and phase distribution, commercial optimization has not yet benefited from accurate sub-surface modeling; therefore, model integration can effectively permit employment of two-phase simulation so as to selectively enhance and assist gas production from gas-rich oil reserves and relatively depleted oil wells.

The concept of integrated modeling and optimization of oil and gas production treats oil reservoirs, wells and surface facilities as a single (albeit multiscale) system [4]: most of all, it focuses on computing accurate reservoir and well state variable profiles. The upper-level optimization directly benefits from low-level two-phase simulation of oil and gas flow, yielding flow control valve settings and production resource allocations. The components of this system are tightly interconnected (well operation, allocation of wells to headers/manifolds, gas lift allocation, control of unstable gas lift wells). Explicit two-phase 3D flow simulation via a dynamic reservoir simulator (ECLIPSE®) is combined with an equation-oriented process optimizer (gPROMS®), to enable integrated modeling and optimization of a case of reasonable size and complexity. An asynchronous fashion is employed: the first step is the calculation of state variable profiles for the combined production system (reservoir+wells) via ECLIPSE®. These dynamic state variable profiles (pressure, oil+gas composition and velocities) are a lot more accurate than the piecewise linear approximations previously used, serving as accurate initial conditions for a dynamic optimization model (gPROMS®). Considering the oil and gas pressure drop evolution within the reservoir and along the wells, we solve single-period or multi-period dynamic optimization problems that yield superior optima, because piecewise linear pressure underestimation is avoided.

REFERENCES

1. Fang, W.Y., Lo, K.K., ?A generalized well management scheme for reservoir simulation?, Paper SPE 29124, Society of Petroleum Engineers (1996).

2. Kosmidis, V.D., Perkins, J.D., Pistikopoulos, E.N., ?Optimization of well oil rate allocations in petroleum fields?, Ind. Eng. Chem. Res. 43(14): 3513-3527 (2004).

3. Kosmidis, V.D., Perkins, J.D., Pistikopoulos, E.N., ?A mixed integer optimization formulation for the well scheduling problem on petroleum fields?, Comput. Chem. Eng. 29(7): 1523-1541 (2005).

4. Gerogiorgis, D.I., et al., "Dynamic Oil and Gas Production Systems Optimization via Explicit Reservoir and Well Multiphase Flow Simulation", Proceedings of the Joint International ESCAPE-16/PSE 2006 Conference, Garmisch-Partenkirchen, Germany (2006).

5. EU Marie Curie PRISM Research Training Network: ?Towards Knowledge-Based Processing Systems? (2004).