Skip to main content
placeholder image

Evolutionary design of marine riser systems

Conference Paper


Abstract


  • It is said that nature attempts to solve the 'problem of life' through evolution. 'Solutions' are proposed which are then tested in the world around us. Natural selection ensures that the best characteristics are inherited by subsequent generations. This paper gives details of a new method of automated riser design and optimization using techniques based on evolutionary theory. Genetic algorithms are a subset of evolutionary computation which rely on natural selection to evolve good designs. A given design (e.g. riser system) is represented by a 'genome' in which the design variables are encoded in the form of'genes'. The optimization software interfaces with an industry standard marine simulation package for design evaluation which gives rise to a great deal of flexibility and few limitations on model complexity. The software can be used for whole system design including multiple risers or as an assistant for the optimization of specific design variables. The software is capable of evaluating systems using both static and dynamic simulations for any number of loadcase scenarios. The design of a titanium catenary gas export riser intended for Statoil's Kristin semi-submersible platform in the North Sea is used to illustrate the method. The design created by the evolutionary software is a significant improvement on the design created using a traditional approach. The results demonstrate improvements in dynamic response together with a reduction in the riser bill of materials cost of approximately one-third, whilst the time spent on design was reduced by nearly an order of magnitude.

Publication Date


  • 2004

Citation


  • Cunliffe, N., McCarthy, T., Baxter, C., & Trim, A. (2004). Evolutionary design of marine riser systems. In Proceedings of the International Conference on Offshore Mechanics and Arctic Engineering - OMAE Vol. 1 B (pp. 733-739). doi:10.1115/OMAE2004-51415

Scopus Eid


  • 2-s2.0-11244340601

Web Of Science Accession Number


Start Page


  • 733

End Page


  • 739

Volume


  • 1 B

Issue


Place Of Publication


Abstract


  • It is said that nature attempts to solve the 'problem of life' through evolution. 'Solutions' are proposed which are then tested in the world around us. Natural selection ensures that the best characteristics are inherited by subsequent generations. This paper gives details of a new method of automated riser design and optimization using techniques based on evolutionary theory. Genetic algorithms are a subset of evolutionary computation which rely on natural selection to evolve good designs. A given design (e.g. riser system) is represented by a 'genome' in which the design variables are encoded in the form of'genes'. The optimization software interfaces with an industry standard marine simulation package for design evaluation which gives rise to a great deal of flexibility and few limitations on model complexity. The software can be used for whole system design including multiple risers or as an assistant for the optimization of specific design variables. The software is capable of evaluating systems using both static and dynamic simulations for any number of loadcase scenarios. The design of a titanium catenary gas export riser intended for Statoil's Kristin semi-submersible platform in the North Sea is used to illustrate the method. The design created by the evolutionary software is a significant improvement on the design created using a traditional approach. The results demonstrate improvements in dynamic response together with a reduction in the riser bill of materials cost of approximately one-third, whilst the time spent on design was reduced by nearly an order of magnitude.

Publication Date


  • 2004

Citation


  • Cunliffe, N., McCarthy, T., Baxter, C., & Trim, A. (2004). Evolutionary design of marine riser systems. In Proceedings of the International Conference on Offshore Mechanics and Arctic Engineering - OMAE Vol. 1 B (pp. 733-739). doi:10.1115/OMAE2004-51415

Scopus Eid


  • 2-s2.0-11244340601

Web Of Science Accession Number


Start Page


  • 733

End Page


  • 739

Volume


  • 1 B

Issue


Place Of Publication