Simulation of dynamic ice indentation failure
Structural vibration caused by ice crushing is an important phenomenon which has to be taken account in the design of structures. The crushing failure mode and rate can often be the cause of the vibration. Beside the dynamic properties of structure the velocity of ice is known to be important variab...
| Main Authors: | , |
|---|---|
| Format: | Other Non-Article Part of Journal/Newspaper |
| Language: | English |
| Published: |
Aalto University
2013
|
| Subjects: | |
| Online Access: | https://cris.vtt.fi/en/publications/82748eaf-b844-4063-af6b-7fdd9e44836c http://toc.proceedings.com/22333webtoc.pdf http://www.poac.com/Papers/2013/pdf/POAC13_142.pdf |
| Summary: | Structural vibration caused by ice crushing is an important phenomenon which has to be taken account in the design of structures. The crushing failure mode and rate can often be the cause of the vibration. Beside the dynamic properties of structure the velocity of ice is known to be important variable in the process. Simulation of crushing is a challenging task. During the continuous crushing process initially intact ice is broken into flakes and fragments of different sizes. The interaction of fragments with structure and intact ice must be considered in the analysis as previous failures affect subsequent failures. The cohesive surface methodology is known to be suitable for the numerical analysis of fragmentation. The cohesive surface methodology is applied in this paper. The methodology is based on inserting possible fracture planes between all elements of the simulation model. This limits the cracking to element boundaries but allows fragmentation and subsequent interaction of fragments. A new rate dependent stress-separation law has been proposed and implemented into the explicit solver of Abaqus software. The applied method has been verified by simulating structure-ice interaction with varying ice velocity. The simulations in this paper focus on determining velocity effects of ice crushing. The simulated results have been compared with the existing laboratory experiments. The failure modes obtained in the simulation were similar to the modes obtained in the experimental tests. In the simulation the contact pressure was found to be not evenly distributed but concentrated into spots. The location of the hot spot was varying during the simulations as it is in the experiments. The highest load was obtained at the beginning of the simulation, similar as in the experiments. After initial failure the results show fluctuations in the contact force. The simulation results are promising. |
|---|