Ice Induced Frequency Lock-In Vibrations - Converging Towards Consensus

A moving ice field or ice feature hitting a fixed offshore structure will cause different ice load actions as ice can be breaking in different ways depending on ice velocity. At low ice velocities loads are pseudo static or intermittent load fluctuations, and at high ice velocities exhibit random variations. In between is a velocity range where the most severe dynamic ice loading on structure may occur: the frequency lock-in range where ice failures repeat continuously close to a natural frequency in a resonant type of a loading. The capricious frequency lock-in vibrations were encountered already 50 years ago, but even now, the true understanding of the ice-structure interaction based on physical and mechanical properties of both the ice and the structure are not fully understood. Recent advances in scale model testing and numerical analysis of measurement results provide deeper insight on what is really happening in dynamic ice-structure interaction. This development provides a refined understanding on the phenomena during ice failure process against the structure, and how the structure is responding. The final proof is still ahead: acquiring dedicated full-scale data to verify the findings of scale-model tests and the predictions of numerical analysis. INTRODUCTION Moving ice fields initiated severe structural vibrations to the first Cook Inlet offshore structures in Alaska over 50 years ago. These events were investigated scientifically and results published, Peyton 1966. Later similar dynamic ice-structure interactions occurred at different locations around the world. However, only few full-scale measurement data sets all with limited instrumentation have been captured, e.g. Peyton 1965, Määttänen 1975, Xu et al. 1981. To understand the physics of dynamic ice-structure interaction many scale model tests have been carried through in ice tanks. However, most of these measurements have had only a single degree of freedom vibration model to present real structures with a plethora of degrees of freedom. As ice failure during dynamic ice structure interaction occurs very fast including many different ice load frequency components it is obvious that a SDOF model cannot imitate the true ice-structure interaction. Hence ice researchers have developed handicapped theoretical models that are based on these measured truncated data sets. The true coupling to the underlying physical and mechanical principles during dynamic ice-structure interaction needs more comprehensive test set-ups. Various and even strongly contradicting models and explanations have been presented, from simple mechanical models to nonlinear self-excited theoretical vibration models. Now, after various scale model tests, limited full-scale measurement data sets, and with improved data analysis methods, a cohesive understanding of the ice-induced vibration is emerging. This paper is an extension to my earlier papers, Määttänen, 2014a and b, describing the history of dynamic ice structure interaction during 50 years. Here the perspective POAC’15 Trondheim, Norway Proceedings of the 23 International Conference on Port and Ocean Engineering under Arctic Conditions June 14-18, 2015 Trondheim, Norway