Experimental and phenomenological investigation of dynamic positioning in managed ice

This paper investigates the dynamic positioning of offshore vessels in managed ice conditions using a model-scale dataset from the large ice tank of the Hamburg Ship Model Basin. Experimental data obtained from the European research and development project DYPIC (DYnamic Positioning in ICe) are analyzed to determine the governing signal characteristics of ice loads acting on a drillship model in various managed ice conditions. The results indicate that the mean load level is strongly dependent on the oblique angle but independent of the relative velocity between the vessel and the ice (when it is below 0.51 m/s in full-scale). Furthermore, it is found that the managed ice cover characteristics (namely, the ice concentration, ice thickness, and floe size distribution) impact both the mean load level and the signal variation, leading to significant and rapid transients in the global load signal. These findings are investigated from a phenomenological perspective, and it is argued that ice floe contact networks and accumulated ice mass are responsible for the observed signal dynamics. Finally, both load signal and phenomenological analyses are used to discuss the implications of managed ice on conventional dynamic positioning control systems. It is shown that several core elements of the system are affected and require attention. Improved design considerations are proposed, but further work is required to implement and test the new concepts.