Shell Appomattox Model-Based Operations from Design to Production: A Game Changer in GoM Deepwater Operation

Abstract

Driven by lower-for-longer cost of operations, companies continuously look for effective ways to increase efficiency and support smarter decision-making. This paper will show how large operations such as Appomattox can use an integrated dynamic simulation-based solution throughout the project lifecycle to aid in design verification, operator training, startup support, and real-time surveillance. The recommendations and findings in this paper can be applied to similar project implementation efforts elsewhere in the oil and gas industry. Appomattox is a four-column semi-submersible production platform located 80 miles off the coast of Louisiana in approximately 7,200 ft of water and features a subsea system with six drill centers, 15 production wells, and five injection wells. Appomattox produces from two deep-water fields – Appomattox and Vicksburg, with potential for asphaltene precipitation and scale formation due to high reservoir temperature and sour service. The Appomattox project team decided in the design phase that dynamic simulation would be a useful tool to assist with operator training due to the complexity of the distributed control system (DCS) and the need for operators to gain familiarity with the combined cycle steam system, a first in the Gulf of Mexico. A multi-purpose dynamic simulator (MPDS) was developed that integrates high-fidelity subsea and topsides models together with the Appomattox control system to provide an environment where one can operate or test without the risk of upsetting the actual process. Prior to integration with the control system, the MPDS was used for engineering studies that resulted in design changes well before first oil, generating significant cost savings when compared to finding these issues at startup. Once integrated, it was able to be used for operating procedure validation and multiple rounds of operator training prior to first oil. Other applications have included controller loop tuning – which generated an initial set of tuning parameters for use at startup, and leak detection algorithm testing by introducing subsea leaks of various sizes. The MPDS currently resides in Shell's Remote Operations Center for production support. The MPDS model has been cloned to be used as a Real-time Surveillance System (RTS), running in parallel with the live facility and connected via Open Platform Communications (OPC) to its historian. The Appomattox RTS sits in Shell's Azure cloud, accessible to Shell users from anywhere. Current RTS applications include virtual multiphase metering, blockage detection in wells/flowlines due to asphaltene/scale deposition, and a performance monitoring system for topsides process coolers designed to predict exchanger fouling/plugging. All key calculated values are sent to Appomattox's historian where they are visible on surveillance screens.