Stephen Bruneau

Hyper-Real-Time Ice Simulation and Modeling Using GPGPU

This paper describes the design of an efficient parallel implementation of an ice simulator that simulates the behaviour of a ship operating in pack ice. The main idea of the method is to treat ice as a set of discrete objects with very simple properties, and to model the system mechanics mainly as a set of discrete contact and failure events. In this way it becomes possible to parallelize the problem, so that a very large number of ice floes can be modeled. This approach is called the Ice Event Mechanics Modeling (IEMM) method which builds a system solution from a large set of discrete events occurring between a large set of discrete objects. The simulator is developed using the NVIDIA Compute Unified Device Architecture (CUDA). This paper also describes the execution of experiments to evaluate the performance of the simulator and to validate the numerical modeling of ship operations in pack ice. Our results show speed up of 11 times, reducing simulation time for a large ice field (9,801 floes) from over 2 hours to about 12 minutes.

EXPERIMENTAL INVESTIGATION OF COMPRESSIVE FAILURE OF TRUNCATED CONICAL ICE SPECIMENS

In total, twenty-eight (28) small-scale ice indentation tests have been carried out to study the compressive failure of polycrystalline ice during indentation and to explore the link between various parameters that influence the ice failure processes, using ice specimens having a truncated conical geometry. Taper angle, temperature, indentation rate, indenter shape and grain size are considered as controlled variables in this research program. For the experiments, three geometric configurations (with taper angles of 13°, 21°, 30°) have been used, conducted at temperatures of -10°C and -5°C. Indentation rates of 0.1 mm/s, 1 mm/s and 10 mm/s have been considered using two indenter shapes (a flat plate and a spherical indenter). Two grain size ranges were considered for these tests. The total force and pressure were found to show dependencies on the indentation rate. The force becomes higher and failure process changes from brittle to ductile as indentation rate decreases. For example, in case of the 21o taper angle ice sample, maximum ice loads were 20 kN and 145 kN and peak pressures were 8 MPa and 18 MPa for indentation speeds of 10 mm/s and 0.1 mm/s respectively. The total force also depends on the taper angle of ice sample. The loads increase as the ice samples become flatter. So, the 13° ice sample was stronger than the 30° ice sample. Different shaped indenters also observed to have distinct experimental outputs. Tests that were done using the spherical indenter show lower forces than the tests that were done using the flat indenter. Effects of temperature reveal that the warm tests show a greater tendency to ductile failure than cold tests having same parameters. The ice samples with smaller ice seeds need more force to fail compared to ice samples with bigger ice seeds. To observe the microstructural modification, horizontal and vertical thin-sections of the damaged ice adjacent to the indenter have been examined. Ice particles were collected from the testing area following each experiment to observe the influence of different factors on the particle size distributions. The effect of each variable on observed failure processes and associated loads are presented in the thesis.

EXPERIMENTAL STUDY OF DYNAMICS DURING CRUSHING OF FRESHWATER TRUNCATED CONICAL ICE SPECIMENS

Ice crushing dynamics in ice structure interactions can result in hazardous vibrations and potentially damaging loads on offshore structures. Ice cone crushing experiments were conducted in the lab to characterize loading and dynamics processes for compressive failure. The indentation rate, temperature and shape of the ice specimens were varied in control tests so that the sensitivity of the resultant dynamic ice load frequency and amplitude could be determined. The results indicate that all control variables had a marked effect on both the frequency and amplitude of load fluctuations. Indentation rates varying from 0.1 mm/s to 10 mm/s and ice taper angles from 13° to 30° had drastic effects. The effects of temperature also demonstrated variations in force, pressure and dynamic behavior. In addition to load measurements, video was used to observe failure mechanisms and in particular spalling and crushing. In the present paper observations are described, though a thorough quantitative assessment has been published elsewhere. Tactile pressure sensors were also used in the experiments, allowing for the correlation of loads and processes to pressure distributions. Finally, the forensic examination of crushed specimens also provided insights into the behavior of ice under various compressive failure scenarios. On the surfaces of intact specimens and revealed within through cross-polarized views of thin sections were signs of ice damage and recrystallization zones of varying extents. The effects of the variables on the dynamic processes and failure behaviors are discussed.Copyright

The Influence of External Boundary Conditions on Ice Loads in Ice-Structure Interactions

Design ice loads are generally derived from field measurements or laboratory experiments. The latter commonly neglect the circumstance that most ice-structure interactions occur underwater, despite the fact that studies report higher ice loads if water is present. Other than a few studies on ice extrusion processes, most investigations on ice loads also do not specifically consider the presence of snow or granular ice at the ice-structure interface. To elucidate the influence of water, snow and crushed ice, as external boundary conditions, on ice load magnitude, 71 small-scale laboratory tests were carried out. Testing involved a hydraulic material testing system (MTS machine) located in a cold room at -7°C. Ice specimens were conical shaped with 25 cm in diameter and with 20° and 30° cone angles. Those were impacted with a flat indentation plate at 1 mm/s, 10mm/s and 100 mm/s indentation rates. Timepenetration and time-force histories from the MTS machine, as well as qualitative contact area and local pressure measurements from tactile pressure sensors were collected. Tests were also recorded with a high speed camera and monitored with still photos. The effect of submergence was most evident at high indentation rate, yielding high ice loads. Snow and granular ice caused comparably high ice loads at the high indentation rate. Moreover, the snow and granular ice conditions also dramatically increased loads at the low indentation rate. In all cases, higher ice loads were associated with increased effective contact areas. INTRODUCTION Icebergs still pose a significant risk of damage to marine structures. In 2000, Hill [1] introduced an iceberg collision database with information on environmental conditions, and damage severity. The database comprises 670 events between 1810 and 2004 involving fishing boats, passenger ships, tankers, bulk carriers and freighters. In just above one quarter of the events, the vessel sank or had to be abandoned. The numbers of accidents and extent of damage reveal the need for measures to reduce risks and damage. Field studies that naturally involve ice impacts in water are most often conducted with instrumented ships. For instance, Masterson and Frederking [2] examined local pressures and forces on icebreakers that rammed ice floes. Later, in 2001, the Canadian Coast Guard Ship Terry Fox was equipped with strain gauges and 178 ice impacts with bergy bits (up to 20,000 t) were accomplished (Ritch et al. [3]). However, those studies do not allow a comparison with loads derived under dry circumstances to directly assess the effect of the water. Most other field indentation tests focus on the influence of different indenter shapes (e.g. Frederking et al. [4], Masterson et al. [5], Kennedy et al. [6]) but not on external boundary conditions. This is despite the fact that ice-structure interactions most likely occur underwater, or at least partially underwater. Ice strength information is often derived from laboratory dry tests and laboratory experiments are still essential to investigate the processes involved in ice-structure interactions. There are only a few publications that address the influence of submergence during an ice impact. For example a layer of spray water was found to yield higher ice loads compared to a dry impact (Varsta [7]). A first approach was taken in a recent laboratory study (Sopper et al. [8]) on ice impacts that provides clear evidence that submergence significantly influences ice loads, particularly at high indentation rates. Furthermore, little information exists on the difference that snow or granular ice at the ice-structure interface cause in ice load magnitude, and how this compares to other external boundary conditions. Most studies focus on ice