Author(s): Ekaterina Kharik; Varvara Roubtsova; Brian Morse; Mario Fafard; Alain Cote; Andre Taras
Keywords: No Keywords
Abstract: Assessing ice loads on dams is important to design safe structures in countries with cold climates (Canada, USA, etc. ). The magnitude of static ice loading events depends primarily on the amount and rate of the ice cover’s thermal expansion, ice type, reservoir water level fluctuations and crack patterns in the ice cover in front of a dam. In addition, the ultimate load could be influenced by the initial stress conditions within the ice sheet that exist due to a complex time-history of meteorological processes and reservoir operations. This paper presents the simulated impact of these initial stresses on a loading event with specific reference to that measured at the Barrett Chute dam, ON, Canada (Taras et al., 2011). Using a finite element model (FEM) with ANSYS, the impact of different initial conditions is explored. The FEM is based on the work of Sinha (1978), generalized by Zhan (1993) and calibrated to Frederking’s (1977) biaxial experiments on T1 snow ice and S2 columnar ice. This model takes into account the differences in material behavior of two ice types. Thermalstructural calculations are conducted to analyze an event measured in the field. The simulations explain how the presence of initial stresses in the ice cover influences the time-history of the loading event and the maximum simulated load. This work is important as, generally, we do not know the initial state of the ice at the beginning of design events. It provides an estimate of the confidence with which design loads can be calculated. The FEM simulations demonstrated that, should the ice sheet experience a cold snap or a very-cold snap prior to the loading event, the initial stresses in the sheet can lead to an increase of 4% to 10% respectively in the subsequent warming event’s peak load.