(418g) Stochastic Pooling Problem for Natural Gas Production Network Design and Operation Under Uncertainty

Natural gas is a vital component of world's energy supply and it is more efficient and less carbon-intensive than other fossil fuels. With world oil prices assumed to return to previous high levels after 2012 and remain high afterwards, consumers are expected to choose comparatively less expensive natural gas for their energy needs whenever possible [1]. The natural gas production networks studied in this paper include gas wells, pipelines, gas production platforms, riser platforms, simple mixing and splitting units and gas terminals (such as liquefied natural gas plants or dry gas processing and distribution terminals) that process and supply gas to markets. The integrated design and operation of such networks determines the optimal network design decisions and the gas flows during operation that can achieve the best expected profitability while ensuring the minimum supply of gas to markets and satisfaction of product specific requirements.

Although the quality (i.e., composition) of gas produced by different reservoirs can vary over large ranges, different gas flows are usually sent through the pipeline networks with little or no processing, to the gas terminals, where they must satisfy strict quality specifications. Therefore, in the integrated design and operation of natural gas production networks, the qualities of the gas flows must be tracked throughout the entire network to guarantee the satisfaction of quality specifications. However, this issue has not been addressed in the literature and it is not taken into account systematically by the design procedures used in industrial practice. In addition, large uncertainties exist in various parts of the network to be designed, such as reservoir qualities, customer demands, etc., and the importance of addressing uncertainty in oil and gas production system design has been well recognized in the literature (e.g., [2][3]).

The major contribution of this paper is to present a stochastic pooling problem formulation that systematically addresses product quality and uncertainty in the integrated design and operation of natural gas production networks, where the qualities of gas flows in the network are described with a pooling model [4][5] and the uncertainty in the network is handled in a multi-scenario stochastic recourse approach. Different economic objectives, including annualized profit, net present value and internal rate of return, are considered for the stochastic pooling problem, and multi-objective problems involving economic and other system specific objectives are tackled via a hierarchical optimization approach.

The advantages of the proposed formulation are demonstrated with cases studies involving one example system from Haverly's pooling problem [4][5] and a real industrial system, the Sarawak Gas Production System [6]. The stochastic pooling problem is a nonconvex mixed-integer nonlinear programming (MINLP) problem which is solved with a recently developed rigorous decomposition method. These results demonstrate the viability of the proposed stochastic pooling problem formulation for real industrial problems with large numbers of scenarios.

References

[1] International Energy Outlook 2009. U.S. Energy Information Administration, 2009.

[2] Goel V, Grossmann IE. A stochastic programming approach to planning of offshore gas field developments under uncertainty in reserves. Computers and Chemical Engineering. 2004;28:1409?1429.

[3] Tarhan B, Grossmann IE, Goel V. Stochastic programming approach for the planning of offshore oil or gas field infrastructure under decision-dependent uncertainty. Industrial and Engineering Chemistry Research. 2009;48:3078?3097.

[4] Haverly CA. Studies of the behaviour of recursion for the pooling problem. ACM SIGMAP Bulletin. 1978;25:29?32.

[5] Haverly CA. Behaviour of recursion model - more studies. ACM SIGMAP Bulletin.1979;26:22?28.

[6] Selot A, Kuok LK, Robinson M, Mason TL, Barton PI. A short-term operational planning model for natural gas production systems. AIChE Journal. 2008;54:495?515.