Artificial neural network development for stress analysis of steel catenary risers- Sensitivity study and approximation of static stress range | www.esteco.com

Artificial neural network development for stress analysis of steel catenary risers- Sensitivity study and approximation of static stress range

Author(s): 
Lucile M. Quéau, Mehrdad Kimiaei, Mark F. Randolph (University of Western Australia)

CHALLENGE - Fatigue design of steel catenary risers (SCRs) is an important challenge especially in the touchdown zone (TDZ) - the area of dynamic riser-soil interaction. Numerous parameters affect the fatigue damage in the TDZ, including those pertaining to riser motions, riser characteristics and soil properties. This paper aims to test the robustness of previous research and extend the ranges of the input parameters for SCR systems under static loading, by means of numerical simulations.

SOLUTION - The numerical models were created and post-processed by means of an in-house interface developed between the optimization software modeFRONTIER and the marine analysis software OrcaFlex through the programming language Python. DoE techniques were used to assist the selection of cases forming the database in the aim of studying the relative effects of the input dimensionless groups and their interactions. This was followed by training a response surface and defining a suitable approximation of Max TDZ/E using ANNs.

 

 

 

 

 

 

 

 

BENEFITS - The framework developed in this paper provides guidance to assist preliminary fatigue design of SCRs. Numerical simulations, using modeFRONTIER combined with DoE methods and ANNs were demonstrated in SCR fatigue design. An approximation using a series of artificial neural networks is presented; it successfully approximates over 99% of the cases of the database with an accuracy of ±5%.