Ship Motion Prediction for the Ampelmann System
P.M.H. van der Steen
R. Babuska – Mentor
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
During rough wave climates, vessel motions prevent people to be transferred safely form and to an offshore structure. The Ampelmann system is a ship-based, self stabilizing platform that actively compensates all vessel motions to make acces to offshore structures safe, easy and fast. The Ampelmann system uses a Steward platform to compensate ship motions. The platform consists of a rigid base frame and a rigid top frame connected by six hydraulic actuators in parallel. The system actively compensates for 5 DoF (surge, sway, heave, roll and pitch), which are measured by an Octans motion sensor. The aim of this master thesis research is to find a algorithm which predicts real-time short-term vessel motions. A pure time series forecasting approach was followed and both linear as non-linear models purposed. Real vessel motion data is used to compare the models. The linear AR and ARMA give stable results, where the AR model surpasses the ARMA model in performance. As non-linear model a Wavelet Neural Network is trained, which improves the prediction compared to the AR model, when the data is re-sampled. The prediction model could be used to improve the compensation done by the Ampelmann system in several manners. For example the delays in the system could be decreased or MPC could be applied.