BP has reported successful deployment of Multivariable Predictive Controllers (MPC) on its Gulf of Mexico deepwater Marlin platform. BP is using Honeywell’s ‘Profit Controller’ MPC system to maximize Marlin production, taking account of a variety of changing operating and weather conditions.
MPCs and Advanced Process Control (APC) solutions are used to manage operations with complex process constraints, where there are significant interactions between variables. BP first deployed an MPC on its Norwegian North Sea Ula platform where a 2% increase in production was achieved.
A key element of the Marlin deployment was the involvement of facilities operators who provided key information on how best the facility operated in both ideal and non-ideal conditions. This let Honeywell’s designers identify the production system’s key operating constraints. The primary compressor system was selected as the best candidate for APC control.
Honeywell’s Experion Process Knowledge System (PKS) was interfaced to third party controllers and a data historian, all running under Profit Controller. Initially, the optimizer was run very conservatively, but after testing, constraints were gradually relaxed until the physical limit of the process was reached.
Substantial engineering effort was required to research and get approval for such a change in operational conditions. The team worked with the rotating equipment manufacturers, Compressor Control Corporation, Honeywell and BP engineering and operations staff to define the optimal operating envelope. The results were dramatic. The ability to use the maximum capacity of the compressor upped production by 4%. BP estimated that the project repaid the extra investment in under three weeks.
This article originally appeared in Oil IT Journal 2005 Issue # 6.
For more information or to comment on this topic email here.