Artificial neural networks meta-models for prediction of vessel offsets and mooring lines tensions of floating production systems.
Edivaldo Delgado1; Aline De Pina1; Bruno Monteiro1; Carl Albrecht1; Breno Jacob1
1 Federal University of Rio de Janeiro (UFRJ)
doi:10.20906/CPS/CILAMCE2015-0360
Resumo
Floating production systems (FPS) for oil exploitation are subject to the action of environmental loads such as wave, wind and current in different incidence directions and varying intensities. The mooring system is responsible for ensuring the integrity of the entire system and maintaining operations safe throughout its useful life. One way of evaluating the efficiency of a mooring system, by means of a numerical tool based on finite elements, is to apply extreme loads in different directions and check the resulting offsets of the floating platform as well as tensions in lines. However, this procedure requires the execution of static and dynamic analyses that have a high computational cost. Therefore, this paper presents an application of artificial neural networks (ANN) to estimate the offsets of the vessel and tensions in the mooring lines by taking as input the main parameters that define the configuration of a mooring system, namely the radii and angles, pretension and material type of the lines. The performance of ANN meta-models were evaluated using statistical tools (RMSE) in terms of accuracy and computational time, by applying to a FPS representative of those currently considered for ultra-deep water scenarios in Southeastern Brazil pre-salt fields.
Palavras-chave: ANN; Mooring System; Meta-models; Offshore Systems