Send comments on this issue and contributions for next issue to iornewsletter@senergyltd.com by 30th April 2003.
Stochastic Integrated Asset Management Helps Choose Best Field Development Strategy
Heikki Jutila
Introduction
A number of commercial software packages have recently become available to perform integrated asset management (GAP, ReO, PipeSim etc.). They are all well developed and provide a robust platform to perform deterministic studies. However, they fall short when the uncertainty in reservoir models, changes in operational parameters and so on have to be modelled. Experience gained by Ingen in performing its day-to-day work has led to the development of a process that encompasses both the rigorous network modelling and the uncertainties prevailing in our industry. The process is illustrated with a case study: an example of the application of this process in the early development of an oil field.
The case study is an offshore subsea development. The reason for this is that in this environment total integration of all disciplines is important (integrate-to-optimise). Traditionally all disciplines have worked in splendid isolation; the flow of information has been in one direction, with the next discipline assuming that the model received from the previous discipline is always correct. In marginal projects and challenging environments, integration of and cross-fertilisation between the disciplines is probably the most important enabler. Any process that promotes this behaviour is of benefit. Ingen has combined this process into a workflow that enables the integration of all disciplines input in a collaborative decision making environment.
Example of Application of RAVE™ to Field Development Scenario Selection
The example field is a satellite of an existing field. One end of the field (Near reservoir) can be reached from the host with extended reach drilling, the other end (Far reservoir) is beyond the reach of platform drilling and it would need a subsea tie in. The development is marginal in its STOIIP and the issue is complicated further by the combination of a low reservoir temperature and a high wax appearance temperature. This means that flowrates need to be maintained in order to keep flowing temperatures as high as possible and that the flowline insulation requires a low U-value.
The reservoir description is summarised as a structure with about 60MMbbl STOIIP; 2/3 in the Near and 1/3 third in the Far; there is some structural uncertainty in the area between the two accumulations. The accumulations might be in partial contact either directly or through the aquifer. The reservoir quality is generally good with reasonable productivities to be expected. The reservoir fluid is undersaturated oil with high wax appearance temperature, 35˚C with a reservoir temperature of 72˚C. The initial GOR is 400 scf/bbl, μoi 1.1 cP, Boi 1.2, Bob 1.3, pi 4,000 psia, pb 1,500 psi. The permeability reflects the high porosity of 35%, the relative permeabilities are equally favourable.
The Operator has to determine the best option for the development of the field. Does he drill an additional producer in Near or an injector, or is the best choice to drill an injector/producer pair in Far. What if there is communication between the two reservoir and what if the aquifer is larger than expected? There is still large uncertainty in the STOIIP estimates for the accumulations and the drilling costs are also subject to considerable variation.
Incorporating all these uncertainties in a set of mechanistic reservoir simulations and economics would be very time consuming and it is very likely that the decision based on these will be "precisely wrong".
RAVE™ is a pressure network model that describes the whole oil/gas production system; starting with the inputs (injection and reservoirs/aquifers) going through the wells and pipelines to the sales point. Figure 1 is a screen dump of the network.
Figure 1: Screen dump of network (Click for larger view)
The flow is from "left to right" with the injection system and injection wells at the left. The pressure then passes through the reservoirs which also might be sources through the wells/gathering system network to the sales point. Once the network is created and the accuracy of the solution has been checked it is very fast and easy to create a set of decision trees that can be used to evaluate various development scenarios quickly and effectively. The methodology is described in more detail in References 1 and 2. Figure 2 shows a typical decision tree with a number of nodes.
Figure 2: Typical decision tree with a number of nodes (Click for larger
view)
The boxes next to the nodes contain pertinent information to the decisions being made, in this case Total Oil Produced. The highest recovery seems to be in the cases at the bottom right with approximately 20 MMbbl recoveries, the Base Case (do nothing) has a recovery of approximately 6 MMbbl.
The value of the process can be augmented further with the application of probabilistic ranges to the key uncertainties; such as STOIIP, drilling cost, well reliability, aquifer size and strength, and degree of communication. In RAVE™ this is an integrated part of the workflow. Once the deterministic cases have been investigated any of the decisions can be subjected to a thorough stochastic analysis. Figure 3 shows the cumulative probability curves for the total reserves for both the Base Case and the FarInj2003 Case. This graph clearly illustrates the benefits of the chosen development plan and the change in the uncertainty and reserve range associated with each scenario.
Figure 3: Cumulative probability curves for total reserves (Click for
larger view)
The application of Stochastic Integrated Asset Management is now possible and the benefits can be clearly seen even in this simple example. RAVE™ is undergoing continuous development and at the moment the abstraction of the reservoir behaviour and the interaction between aquifers and reservoir compartments is receiving a lot of attention.
References
1. P W Gayton (Ingen), S D Miller (Baker Jardine and Associates), R Napalowski (Veba Oil and Gas Ltd.) "Innovative Development Engineering Techniques", SPE 65202 (2000) (http://www.ingen-ideas.com/html/resources/papers/SPE65202.pdf)
2. A L Robertson (Ingen), P R P Cunningham (Ingen), P W Gayton (Ingen), M K Castell (Chevron UK Limited), N A Menzies (Chevron UK Limited) "A Systems Engineering Approach to Field Development", SPE 71833 (2001) (http://www.ingen-ideas.com/html/resources/papers/SPE71833.pdf)
Disclaimer: The material available on this website is designed to provide general information only. Whilst every effort has been made to ensure that the information provided is accurate, it does not constitute legal or other professional advice.
Please note: The Department of Trade and Industry cannot be held responsible for the contents of any pages referenced by an external link.
