In-time fatigue monitoring using neural networks

This paper presents a new methodology for fatigue monitoring of fixed offshore structures. The accumulated fatigue damage on selected joints, caused by actual loading conditions, will be assessed through a data acquisition and processing system installed on board. Based on measurements of upward jacket displacements and feedforward backpropagation neural networks, the calculation of fatigue damage is performed in-time. This information, along with inspection reports, could be used to improve further inspection and repair of the platform jacket. A simplified structural numerical model takes into account, as much as possible, the dynamical structural conditions of a platform installed on Campos Basin, RJ, Brazil, in order to match the paper's conclusions to the actual behaviour of fixed offshore structures.