Steel Lazy Wave Riser Optimization Using Artificial Intelligence Tool

Abstract

Use of steel lazy wave risers has increased as oil and gas developments are happening in deeper waters or in parts of the world with no pipeline infrastructure. These developments utilize FPSO’s with offloading capabilities necessary for these developments. However, due to more severe motions compared to other floating platforms, traditional catenary form of risers are unsuitable for such developments and this is the reason Steel lazy wave risers (SLWR) are required. SLWRs have shown to have better strength and fatigue performance and lower tensions at the hang-off compared to steel catenary risers. A suitable Lazy-Wave form of the catenary riser is achieved by targeted placement of a custom configured buoyancy section. The strength and fatigue performance of steel lazy wave risers are governed by parameters such as length to start of this buoyancy section, length of the buoyancy section, hang-off angle and the buoyancy factor. Achieving these key performance drivers for a SLWR takes several iterations throughout the design process. In this paper, genetic algorithm which is an artificial intelligence optimization tool has been used to automate the generation of an optimized configuration of a steel lazy wave riser. This will enable the riser designer to speed up the riser design process to achieve the best location, coverage and size of the buoyancy section. The results that the genetic algorithm routine produces is compared to a parametric study of steel lazy wave risers varying the key performance drivers. The parametric analysis uses a regular wave time domain analysis and captures trends of change in strength and fatigue performance with change in steel lazy wave parameters.