Implementation, verification and validation of the multi-surface failure envelope for ice in explicit FEA

• Main Authors: Wang, J., Derradji-Aouat, A.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Curran Associates 2009
• Subjects: Arctic
• Online Access: https://nrc-publications.canada.ca/eng/view/object/?id=f783fe0b-7f9c-4f21-a1f3-0700946d6e2c; https://nrc-publications.canada.ca/fra/voir/objet/?id=f783fe0b-7f9c-4f21-a1f3-0700946d6e2c

The multi-surface failure criterion/model for ice is implemented into an explicit finite element program called LS-DYNA (www.lstc.com). The implementation of the model equations was achieved via the development of a user routine for LS-DYNA. Since ice behavior depends on temperature, strain/load rate and loading direction (loading direction for the case of anisotropic ice), the traditional isotropic and temperature and rate independent failure criterion such as von- Mises and Mohr-Coulomb are not applicable. The multi-surface failure criterion (Derradji-Aouat, 2003) accounts for those effects and is formulated with the parameters of the octahedral shear stress and hydrostatic pressure. In this paper, a 3-step procedure will be presented and discussed to demonstrate successful multisurface failure model implementation in explicit FEA. Step 1 is concerned with the theoretical implementation of the ice multi-surface failure model into the explicit commercial code LSDYNA. Step 2 deals with verification and validation of the FE implementation, this step may be called V&V analysis and its rooted in uncertainty methodologies and statistical analyses. V&V will be based on comparisons between the numerical results and the results from the ice compression tests done by Rist and Murrell (1994) which was one of the data sources for the present failure criterion. Step 3 deals with showing how the implemented model in LS-DYNA predicts actual ice pressure and indentation tests on an ice block using the MTS machine in the cold room (Wells et al., 2008) and are simulated as an example. Discussion regarding the multisurface model, its implementation, numerical results, and model predictions are presented. Peer reviewed: Yes NRC publication: Yes