Abstract

Side-by-side operation of multiple floaters in the ocean environment is very challenging and the operators always prefer a maximum operable weather window, in order to minimise the cost incurred from the downtime. The safety of the gangway connecting the floaters is very crucial and its dynamic response in the ocean environment raises concerns during operations. Therefore, an efficient dynamic positioning system is essential to maintain the floater and ultimately, the gangway response within the desired limits. In this work, a novel dynamic positioning system for floater operating aside another vessel is presented. The system includes an adaptive controller combined an optimised thruster allocation law and with a sea state detector. The adaptive control is achieved by using the barrier Lyapunov function and a predictor-based method in combination with the neural network scheme. The limitations include the saturation of inputs and the forbidden zones due to thruster-thruster or thruster-hull interaction. An optimised allocation for lower fuel consumption, wear and tear of the thruster equipment and to ensure the resultant command in the respective direction of the azimuth thrusters is designed. The optimisation here is a non-convex problem and a locally convex reformulation of second order is implemented. The presence of unknown shielding effect due to nearby vessel in a side-by-side configuration and input time delay is also considered in the development of this thruster allocation law. In order to overcome these effects, a novel sea state detector is designed. The sea state detector can effectively monitor the variation of drift wave-induced force on the vessel and activate the neural network compensator in the controller when a large wave drift force is identified. Simulation studies are conducted to verify the efficiency of this dynamic position system and a demonstration of flotel in side-by-side configuration with a turret moored FPSO is presented for the non-collinear ocean environment.

This content is only available via PDF.
You do not currently have access to this content.