Armed, as most of us are, with perfect 20/20 hindsight, how many times have we looked at a situation and thought, "If only we had done things differently..." Unfortunately this is often the case in oil and gas production. Every day, designers face huge challenges to walk the fine line between over designing their production units, incurring higher CAPEX and OPEX, and under designing them, thus limiting their potential to optimize profitability.

Once the project is on stream, there is still an opportunity to optimize asset performance to make the most of the design base to which the asset is committed, or to remediate the design early enough in the field life to improve project performance. The ideal asset management system would predict reservoir problems before they impact the wellbore, maintain production equipment within its operating window and, importantly, assess short-term productivity opportunities with reference to strategic depletion plans.
In most cases the problem is information - inaccurate information, information containing gaps or discontinuities, irrelevant, untimely information or information that is not information at all, just data.

End-to-end integration is key

A system that is capable of communicating with each member of the design and management team in that specialist's own "language," and one that can propagate the effect of decisions implemented anywhere in the model to the rest of the model, is fundamental to design optimization and daily operation excellence. Even the most sophisticated reservoir models, well models, gathering network models, process facilities models and economics models are not going to achieve desired results unless they can communicate seamlessly with one another at the right time.

With integration, domain specialists can perform concurrent tasks while identifying and resolving conflicts that add costs or impair productivity. Similarly, they can exploit synergies that add efficiency, boost output or extend economic life. Here are a few applications:

• Dynamic modeling of reservoir and surface:
- Network balancing.
- Reservoir coupling.
• Integrated production modeling:
- Short term, tactical operations planning and de-bottlenecking.
- Medium and long term strategic production forecasting.
• Field-wide optimization:
- Allocation of limited resources such as lift-gas, steam injection and ESP power along with associated process facility constraints.
- Field development planning strategy.
• Gas field management:
- Compressor operation and planning strategy.
- Managing contracts and nominations.

The integrated solution, from Schlumberger Information Solutions, is called Avocet Integrated Asset Model. It is being implemented in several stages, with Stage 1 available now. The current implantation enables a tight coupling of the ECLIPSE reservoir simulator with the PIPESIM surface network simulator and a link to the HYSYS process simulator to enable integrated design. Alternatively, a tight coupling between HYSYS and PIPESIM is available to facilitate operational workflows. The next stage of development will couple the reservoir model with the pipeline model with the HYSYS process simulator and economic analysis modules. The integration is being achieved under an alliance between Schlumberger and Aspen Technology, Inc. (AspenTech). Stage 3, planned for delivery in 2006, will provide role-based portfolio visualization for corporate portfolio management and decision-making, and will enable effective management of multiple assets, including partnership deals. The resulting integrated asset model, supporting digital oilfield workflows, is one step closer due to the collaboration of AspenTech and Schlumberger on an end-to-end model management system from reservoir through process. The partnership includes the development of field controllers to support well management workflows and global optimization across the entire asset. Part of the research undertaken in the alliance is how to propagate uncertainty management from the reservoir through the facilities.

How does it work?

To visualize how integrated asset management can simplify a complex development scenario, consider an example where the integrated asset model workflow has been applied to the data for a North Sea "Indigo" field, a prolific black oil reservoir about 12,000-ft (3,659 m) deep. The field was producing 30°API black oil from 17 subsea wells and 2 injectors. Water depth is 245 ft (75 m). Over Indigo's 15-year history, a comprehensive, 300K-cell reservoir model had been constructed using ECLIPSE. On the seabed, wells, flowlines and manifolds were modeled using PIPESIM to characterize the complex subsea production network that culminated in a 31-mile (50-km) multiphase pipeline to the processing platform.

The asset management team is now called to action when a second reservoir, Indigo 2, is discovered about 7.5 miles (12 km) to the south. A compositional field of condensate with associated gas, Indigo 2 development plans call for seven producing wells and two injectors. Economics are marginal and development is expected to take 12 months. Production has to be routed to the same processing platform used by Indigo 1. The question the team needs to answer is, "What's the best way to co-produce the fields?"

Two development options are postulated. Option 1 involves gathering flow from Indigo 2 at a subsea manifold and tying it to the Indigo 1 flowline at its manifold. Option 2 calls for a completely separate flowline and riser system for Indigo 2, with a dedicated first stage separator at the platform.
A rigorous study can be conducted to first determine the cost/benefit implications of each alternative, but equally important, the predicted effect of reservoir and surface interaction when both fields are put on stream. Processing and transportation facilities have constraints, such as hardware limitations and export specifications, that must be included. In some cases, operators wishing to make an all-encompassing design will actually model the constraints to see if a better solution can be obtained by changing the parameters. In the Indigo case, however, it is determined that production models would be limited by facilities constraints, and thus all designs have to take the constraints into account. Other technical considerations include reservoir coupling, or synchronization of the two reservoir models with the gathering network, and combining the two reservoir models with a common surface network.

Not as easy as it looks

Among the simulations performed during the decision-making process are comprehensive comparisons of reservoir coupling, manifold design and impact on production facilities. Each node of the processing train is modeled, allowing comparisons of such items as platform heating and cooling requirements, compression requirements and water treatment and handling requirements. Each of these will differ depending upon the option chosen. Moreover, the results must be entered into an economic field life model to determine the long-term effects on cash flow.

Reservoir coupling performed under Avocet uses ECLIPSE and PIPESIM to integrate the production and transportation network leading up to the platform. The Aspen AssetBuilder surface coupling workflow process takes the production stream where it comes onboard and aggregates individual comparison models as constrained by the system to determine the final effect on the asset's profitability.

In the case of Indigo, subsea tieback networks had to be upgraded for each option. Option 1, tie the production together at the subsea manifold and transport it to the platform using a single pipeline requires a CAPEX expenditure of US $35 million. Option 2, which involves a second pipeline, costs $120 million. Just looking at the cost of the upgrades, one would have an easy choice, but what is the effect on reservoir production?

Here the reservoir coupling model showed that tying Indigo 2 to Indigo 1 at the manifold would cause a major difference in production volumes due to increased back pressure. Option 1 production would be 44,000 bo/d, whereas Option 2 promises 120,000 bo/d. This is largely due to an increase of subsea manifold pressure by up to 120 psi if Option 1 is chosen.

The study is then continued and investment options compared using AspenTech's Icarus process facility costing program. CAPEX requirements for Option 1 totaled $80.7 million, compared to $188.5 million for Option 2. Whereas the subsea differential is 3.5:1 the differential including the platform facilities is only 2.3:1. Taken together, OPEX net present value (calculated at 10% discount rate) plus CAPEX for Option 2 are $300 million higher than Option 1 over the field's expected life. At this point, Option 1 looks pretty good.

But a high fidelity integrated asset management system must do more that compare nuts and bolts. It must have the capability to manage uncertainties and run risk sensitivity models. For example, variables such as commodity prices must be considered and factored-in. Using Merak PEEP in concert with the Avocet Integrated Asset Model the uncertainties can be analyzed. Decision trees, showing P50, P90 and P10 cases can then be output. Most importantly, after tax cash flow can be determined.

And the winner is...

This is where Option 2 becomes the clear choice. Notwithstanding the foregoing individual node analyses, Option 2 is shown to payout in 8 years, the result of its higher sustained production rate. At this time, annual cash flow stabilizes at $110 million.

Principles of integrated asset modeling and management now provided under Avocet were instrumental in the successful facility design for ChevronTexaco's giant Agbami development offshore Nigeria. Details can be found in SPE 90976 (September 2004).

Comments