Summary
The possibility of updating reservoir models with new well information is
important for good reservoir management. The process of drilling a new well
through to update of the static model and to history match the new model is
often a time-consuming process. This paper presents new algorithms that allow
the rapid updating of object-based facies models by further development of
already existing models.
An existing facies realization is adjusted to match new well observations by
changing objects locally or adding/removing objects if required. Parts of the
realization that are not influenced by the new wells are not changed. A local
update of a specified region of the reservoir can be performed, leaving the
rest of the reservoir unchanged or with minimum change because of new
wells.
In this method, the main focus is the algorithm implemented to fulfill well
conditioning. The effect of this algorithm on different object models is
presented through several case studies. These studies show how the local update
consistently includes new information while leaving the rest of the realization
unperturbed, thereby preserving the good history match.
Introduction
Rapid updating of static and dynamic reservoir models is important for
reservoir management. Continual maintenance of history-matched models allows
for right-time decisions to optimize the reservoir performance. The process of
drilling a new well through to updating of the static model and history
matching of the new model is often a time-consuming process. Static reservoir
models and history matches are updated only intermittently, and there is
typically a 1- to 2-year delay between the drilling of a new well and the
generation of a reliable history-matched model that incorporates the new
information.
This paper presents new algorithms that allow rapid updating of static
reservoir models when new wells are drilled. The static-model update is
designed to keep as much of the existing history match as possible by locally
adjusting the existing static model to the new well data. As the name implies,
object models use a set of facies objects to generate a facies realization.
Stochastic object-modeling algorithms have been developed to improve the
representation of facies architectures in complex heterogeneous reservoirs and,
thereby, to obtain more-realistic dynamic behavior of the reservoir models. We
consider the main advantages of object models to be the ability to create
geologically realistic facies elements (objects) and control the interaction
between them, to correlate observations between wells (connectivity)
explicitly, and the possibility of applying intraobject petrophysical
trends.
© 2009. Society of Petroleum Engineers
View full textPDF
(
3,335 KB
)
History
- Original manuscript received:
21 February 2008
- Meeting paper published:
9 June 2008
- Revised manuscript received:
15 September 2008
- Manuscript approved:
22 September 2008
- Published online:
1 June 2009
- Version of record:
1 June 2009
View Cart
