Experiments by Invited Researchers


Ice Interaction with an Arctic Jackup

Project acronym: HyIV-HSVA-04
Name of Group Leader: Dr. Gus Cammaert
User-Project Title: Ice Interaction with an Arctic Jackup
Data Management Report: Report

User-Project Objectives

The type of structure that was modelled in the ice tank tests was similar to the one shown in Fig. 1. This is a special version of a conventional 4-leg jackup designed for extreme environments. The major modification is that while most jackups have legs with open bracing, the Arctic Jackup legs will have 6-8 m diameter cylindrical legs to minimize ice damage and provide high resistance to local and global ice loads.There is a major concern in the industry that in the Arctic this jackup design will only be able operate at the beginning of the ice season, although the “cut-off ice thickness” is not well defined. Ice management will be needed to extend the operating season, and if this can be done successfully such structures will allow economical and safe drilling programs in shallow water depths.The original scientific objectives of the testing program included the following: · For uniform ice conditions where ice load estimates are reasonably reliable according to the available algorithms, carry out a series of “benchmark” tests for those ice conditions which are much more difficult to quantify for load calculation purposes (i.e. broken or managed ice). · For the main series of ice tests to be carried out, investigate the effects of ice management, varying ice floe size, thickness and concentration. · For an additional series of tests, if time and budget permit, carry out tests simulating the effects of managed and unmanaged ice ridges interacting with the jackup legs.

Short description of the work carried out

A series of tests were designed around a simple model of a jack-up structure as shown in Fig. 2. The structure consisted of 219-mm pipe sections connected by bracing. At a model scale of 31.95 to 1, the legs represent column diameters of 7.0 m and leg spacing of 50 m. The tests were performed in 6 testing days, divided over 3 weeks. Three model orientations (0°, 22.5°, 45°) and 2 ice thicknesses (33 mm and 16 mm - or 1.1 m and 0.5 m full scale) were used.Typically three tests were performed each testing day, one level ice test as a benchmark (Fig. 3) and 2 runs with managed ice in concentrations of 8/10ths and 6/10ths. Two velocities were tested in one run. Each of the 4 legs has a 3-axis load cell, and the whole model is suspended on a large 6-component load cell. The measurement results were recorded by a computer on the towing carriage. All data and video files were placed on a central server; the load data was sampled at 50 Hz.The results of the tests were quite successful except for minor delays in the testing schedule due to the time required to obtain the correct ice mechanical properties. Vibrations were somewhat greater than expected and in some cases the failure mechanisms were not anticipated (buckling rather than crushing, for instance). From preliminary checks the loads measured were consistent with the expected results.

Highlights of important research results

Up to now little information has been published on the effects of ice management on ice actions on structures of this type. A MSc graduate student is now analysing the results of the tests. Specific conclusions relating to the efficiency of ice management operations have not yet been determined based on the large volume of load data which has been collected. However, it is anticipated that at least two conference papers and one journal paper will be published documenting the following: · A quantitative comparison of managed and unmanaged ice actions for specific combinations of ice thickness, floe size and floe concentration (Figs. 4 and 5). · A calculation of ice management efficiencies (reduction in ice action for specific thicknesses as a function of changes in ice management strategies). · Comparison of test program results with similar published data for model tests with multi-leg platforms in ice. · Analysis of the recorded dynamic response of the test structure and observed failure mechanisms.Time and budget constraints did not allow simulating the effects of managed and unmanaged ice ridges interacting with the jackup legs.

J.S. Hoving et al.Experiment-based Relations between Level Ice Loads and Managed Ice Loads on an Arctic Jack-up Structure , 2013 Proceedings of the 22nd International Conference on Port and Ocean Engineering under Arctic Conditions (POAC?13), June 9-13. Espoo, Finland, Paper ID: POAC13-122
Rogier VermeulenExperiment-based quantification of managed ice loads on a four-legged structure , 2014 Delft University of Technology - Offshore and Dredging Engineering