Experimental and numerical analysis of ice crushing tests with different shaped ice specimens

• Main Authors: Müller, Franciska, Böhm, Angelo, Herrnring, Hauke, von Bock und Polach, Rüdiger Ulrich Franz, Ehlers, Sören
• Format: Conference Object
• Language: English
• Published: 2023
• Subjects: 620: Engineering and Applied Operations; Arctic; ice covered waters
• Online Access: https://hdl.handle.net/11420/43883

Ships operating in ice covered waters are exposed to the risk of colliding with an iceberg or ice floe. Predicting the loads in ice-structure interaction is crucial for the safety of existing ships and the optimization of the ship structures in the future. The ice geometry and the interaction velocity are two parameters, among others, that have a high influence on the acting forces and pressures. There has not been much research on the influence of the ice geometry in ice-structure interaction tests in the past. This paper analyzes the failure mode, peak forces, and energy of different ice specimen geometries at different interaction velocities in small-scale crushing tests and identifies shapedependent effects. Furthermore, a first attempt is made to numerically simulate these experiments with the Mohr-Coulomb Nodal Split (MCNS) ice model in order to extend the applicability of the model to more general ice-structure interaction problems. Overall 21 different ice shapes were tested in crushing experiments with a hydraulic piston and drop tower experiments and were simulated with the MCNS model. In addition, the influence of the interaction velocity was analyzed for 13 shapes at interaction velocities of 10, 100 and 2000 mm/s. The specimen shapes included among others a cylinder, cones, truncated cones, elliptical, dome-shaped and wedge-shaped. In the experiments, the load, the displacement, and the failure behavior were recorded. Especially the peak force, energy, and failure mode were used for the analysis and comparison with the simulation results. The results show a high dependency of the failure mode and peak forces on the specimen shape, initial contact area, and interaction velocity. The MCNS model is capable to predict trends of the peak load depending on the initial contact area and generated different failure behaviors depending on the ice geometry to some extent. However, there is potentially room for improvement, especially considering the triaxiality and the influence of the interaction velocity.