BP - dynamic modeling mature fields production

Friday, May 11, 2012

BP is implementing software to dynamically model production from mature fields, which can help work out ways to improve production. By Patrick Calvert, BP, and Peter Griffiths, BP

Sustaining production from wells which are reaching maturity has traditionally been a significant challenge for operators looking to maintain and maximize production.

There is a growing need for accurate modelling systems to improve the management of dynamic transients
that often dictate the operability of mature fields.

For example, optimising the dynamic transients can mark the difference between a successful and an unsuccessful well start up and, when in production, optimising flow stability will reduce the need for excessive choking and provide a stable platform to push production closer to capacity.

Fluid conditions experienced during well start-up, liquid loading and slugging are sufficiently removed from the steady state to render steady-state-modelling tools ineffective at simulating the associated behaviour.

As dynamic transients have become more challenging to manage, dynamic modelling has received more recognition amongst petroleum engineers. Software, which has historically been confined to the flow assurance community, has been rapidly
extended to address this growing demand.

Starting with a single project on a platform in the North Sea, BP's dynamic modelling programme has grown to encompass five separate regions. In 2011 dynamic modelling and control accounted for 17.5 mbd of
the company's annualised incremental production.

Initially, BP limited dynamic modelling to low rate wells and flowlines that experienced stability or start up problems. However, the scope was extended to include high rate wells where rapid transients can have significant safety implications.

Figure 1 highlights a series of operating problems that dynamic modelling has proved effective at addressing.


Within figure 1, each challenge has been framed in terms of the superifical liquid and gas velocities, which represent the flow conditions that give rise to the problem. Two areas where BP has seen significant demand for dynamic modelling are well start-up and well clean-out.

Start-up can be problematic in wells with poor pressure support and no artificial lift. If the pressure driving force across the well is insufficient to overcome the maximum hydrostatic head experienced during start-up, any attempt to bean-up the well will fail. The amount of water or dead-oil present within the well will increase for each failed attempt, impacting subsequent start-up attempts. The operating strategy for such wells is critical, as the manifold pressure needs to be sufficiently low to ensure a successful start-up. Dynamic modelling has assisted in providing an understanding of the shut-in pressures required for a successful start-up. By combining commercial modelling software with simple data analytics, an optimum bean-up rate can be determined for each well along with an estimation of the time it takes for the well to unload. To date, the use of well start-up guidelines derived from dynamic modelling has proved critical in the continued production from a number of fields in the North Sea, the Gulf of Mexico and Trinidad.

The ever-increasing rigour required for operational decision-making meant that it was a natural progression to apply dynamic modelling to support drilling activities. Dynamic simulation has helped in the understanding and fine tuning of the steps required to ensure the complete removal of mud in an environment that is constrained by the rate of produced fluid and the total volume of fluids that can be handled during well clean-out operations. Having the in-house capability to model clean-out operations has led to a much stronger collaboration between design teams and the technical contributors. The HAZOP review has been particularly strengthened, and simulated flowing temperatures and pressures have formed a vital feed to this process. Questions and scenarios raised during the HAZOP can be appropriately simulated to provide the information required to complete any remaining actions.

Over the past six years, BP has adopted a three-tiered approach for developing and sustaining dynamic modelling through a process of demonstrating value, crystallizing interest and building regional capability.

The creation of an internal dynamic modelling programme was followed by the establishment of pilots in three North Sea fields through collaboration between the central subsurface technology organization and key staff in the region. The combination of third party applications and in house software enabled BP to create an improved operating strategy for each field, bringing enhanced flow stability and increased production by 3.5 mbd. With few practitioners it was also necessary to express the output of the modelling in a manner that could easily be interpreted by those with no previous experience of dynamic modelling. Visualization was used to convey the key findings derived from the dynamic studies. Particular emphasis was placed on presenting the findings in a format that preserved the 'shelf-life' and recognized the synergy between steady state and dynamic modelling. The latter ensured that dynamic modelling could be tied in with an extensive group of steady state modelling practitioners.

Other influential factors in the success of the pilot included included buy-in from the assets involved and the company as a whole, continuous remote support, easy access to real time operating data, the ability to refine and hone the modelling techniques and a clear understanding of the value that the system added.

The success of the three pilot schemes in the North Sea encouraged the company to appoint a full time modelling specialist for the region. A single modelling programme was developed from the pilots, which could then be coordinated entirely from within the company's North Sea business unit. This initial development in the North Sea, whilst demonstrating success, was also designed to ensure it can now be replicated in many other regions, with application currently underway in several areas including Angola, Trinidad and the Gulf of Mexico.

Over a three-year period, the resource grew from a single specialist to five people with dynamic capability. A North Sea system optimisation team was established to cover the region and support all aspects of real time system analysis. The team is now regarded a good example of how to build a sustainable modelling capability within a company. Its success has led to significant and important internal buy in and requests for further work. The team now has a clear identity, modelling knowledge is shared amongst the individuals and the modelling software and workflows have been standardised.


Conclusion

What was initially a relatively small, regional pilot project for BP has matured into a global initiative across the company. The number of fields involved in the programme has increased leading to to a significant gain in incremental production. Dynamic modelling has enabled BP to increase incremental production 10 fold in the past six years. In 2011, dynamic modelling initiatives led to 17.5 mbd of annualised incremental production.

The company's focus on demonstrating value and building regional capability, through effective application of technology and seeking to develop and standardize best practice, has enabled it to build a permanent capability for sustaining the use of dynamic modelling to support its daily operations.