COST-EFFECTIVE OPTIMAL SOLUTIONS FOR STEEL CATENARY RISER USING ARTIFICIAL NEURAL NETWORK

Over decades, free-hanging steel catenary riser (SCR) has been regarded as a cost-effective and straightforward system for riser solutions in deepwater and ultra-deepwater by National oil companies NOCs, international oil companies IOCs, and independent players. However, the free-hanging SCR comes with its own challenges, which, if not correctly evaluated during design stages, can result in system failures. The two main challenges facing the free-hanging SCR usage are increased self-weight and high fatigue damage, especially at the touchdown and hang-off points. Fatigue damage is the failure resulting from cyclic stress accumulation over a specific duration. The involvement of technologies has improved the design method for these systems over the years. Two of such innovative contributions include the optimisation algorithm and artificial neural network ANN methods. The aim of this research is to develop an effective optimisation algorithm to search for the global optimal weight of the riser configuration while maintaining its structural integrity. The method involves the combined use of ANN and optimisation genetic algorithms (GA) to determine the optimum solution of the SCR. The GA is used because of its capability in handling a variety of complex non-linear optimisation problems to ascertain the global optimum solution and its capacity to self-moderate the number of iterations. At the same time, the ANN method is deployed for its accurate prediction of non-linear responses. The deployed techniques have shown to be promising due to the time-saving and less computational cost compared to integrating GA and time-domain FE models. This method was illustrated using a prospective 10-inch internal diameter SCR installed in a 2000m deepwater offshore field off the oil-rich Niger Delta region, Nigeria. The obtained results have shown a 19.10 per cent reduction in the riser weight and a time reduction of 90.83 per cent.