In this paper, a numerical procedure based on SPH is presented to analyze the failure process of ice. The Drucker-Prager model is implemented into the SPH code to simulate four point bending and uniaxial compression failure of sea ice in a local scale. The procedure for implementing softening elastoplastic model is used in the SPH framework. To validate the model, the numerical results are compared with finite element simulations and experimental results. The good agreement has demonstrated that the presented SPH procedure can be a useful numerical tool for the simulation of failure progress of ice.
With the increasing activities in Arctic regions, the numerical simulation of ice-structure and ice-ship interactions became in the past decade of increasing significance due to its high importance in the design process and better accessibility to test data. In order to simulate ice-structure and ice-ship interactions effectively, it is necessary to have a proper understanding of the ice failure behavior. The bending failure of ice is of high significance for ships in ice due to the inclined contact interfaces with the ice (Valanto, 2001). And while bending as main failure mechanism, the initial contact causes local compressive failure (crushing). The more vertical the contact area or structure becomes, the more compressive features are included in the failure process. This underlines the importance to represent the failure in bending and compression with the same model.
To have a proper understanding of the ice failure behavior, different full-scale tests and model tests were carried out which are readily available in literature (Kujala et al., 1990). However, the experimental results are dispersed because of different testing approaches, the measurements of the test specimens and ice properties. Thus, it is very important to develop a reliable numerical ice model to simulate sea ice failure in bending and compression, especially the current studies on the behavior of sea ice failure are not adequate.
