Summary
In this paper, we propose an optimization framework for maximizing asset
value, both with and without uncertainty. We first present the methodology to
treat a general control optimization in the presence of uncertainty, followed
by a brief section on the optimization algorithms used. We then describe the
field model example used to illustrate the application of the methodology.
Through a systematic analysis of various deterministic and stochastic cases, we
address the various objectives sought. Using net present value (NPV) as a
measure, we also explore the valuation of advanced completions along with the
returns gained from expanding surface gas-handling facilities. The method also
generates an efficient frontier that can be used for risk and decision
analysis. The results clearly demonstrate the value of such a framework for
value maximization in planning both near- and long-term time horizons as well
as providing the necessary foundation for maximizing asset value.
Introduction
We consider an existing infill program for a mature real onshore oil and gas
field with the objective of maximizing asset value. We do so by optimizing a
history-matched reservoir model, and we provide confidence levels under
uncertainty by generating efficient frontiers.
Application of search or optimization algorithms has been the subject of
numerous studies and articles both inside and outside the petroleum
industry.1–14 Following from the work of Raghuraman et al.,1 this paper
considers a real reservoir and attempts to maximize its value by analyzing
various exploitation scenarios.
The paper first describes the main features of the framework: the overall
methodology and different optimization schemes. It then applies the
optimization process to the field example. While maximizing asset value with
and without the presence of uncertainty, the efficient frontier is discussed
and its use for risk management and decision making is demonstrated.
Methodology
The process of optimizing a reservoir, under the assumption that everything
is deterministically known, is relatively straightforward. One may want to
extract the maximum fraction of oil and/or minimize the water production or
maximize the NPV of the oil produced by optimally controlling various
operational variables (e.g., individual-completion flow rates), all the while
accounting for physical constraints (e.g., single-well production or pump/valve
limitations) and economic constraints (e.g., drilling, logging, or stimulation
costs). However, the presence of physical and/or financial uncertainties
elevates the problem of optimization to the level of a risk-management problem.
A framework has been developed that encompasses the necessary elements to
perform reservoir optimization under uncertainty and to provide the risk
analysis necessary for decision making. A detailed description of the process,
with an example on reservoir monitoring and control, is given in Raghuraman et
al.1 Fig. 1 shows a schematic of the algorithm for a problem with
uncertainty.
© 2005. Society of Petroleum Engineers
View full textPDF
(
2,712 KB
)
History
- Original manuscript received:
24 June 2004
- Revised manuscript received:
16 November 2004
- Manuscript approved:
13 December 2004
- Version of record:
15 February 2005
View Cart
