Physical Modeling of Suction Piles in Clay

Abstract

The behavior of suction piles that are installed by suction into soft Speswhite clay was investigated by model tests in a large geotechnical centrifuge. Many different placement and loading tests were carried out in similar soils, thereby providing a reliable method for evaluating the effects of selected parameters on pile response. Techniques and equipment were developed to prepare normally consolidated clay samples with a high degree of repetitiveness and to measure their characteristics in-flight (e.g., on-board movable Cone Penetrometer device – abbreviated CPT in this paper) in order to simulate deepwater offshore soil conditions. The CPT tests performed in-flight indicate that these objectives were achieved with the soil samples tested. Special devices were developed for this program to apply well-controlled vertical and inclined monotonic and cyclic loads. A well-controlled system was necessary to investigate sustained and cyclic loads for simulating environmental loading conditions on anchor pile moorings for floating production and drilling facilities, or on other facilities used in the storage and off-loading of oil and gas. In the test program, the capacity measured during monotonic loading served as a reference value for establishing a scale factor to be used in applying cyclic and sustained loading. A monotonic load test was always carried out before and after each cyclic or sustained load test in order to quantify any change in capacity due to a given loading event. Issues addressed in the program and discussed in this paper include the effect of suction installation methodologies on pile capacity, and the behavior of piles during inclined monotonic loading and cyclic loading due to simulated Gulf of Mexico hurricane events. Other issues were addressed in the program (e.g., simulated loop current events, effect of a sand layer on suction pile penetration) but are not included in this paper due to space limitations. A number of tests were performed for each loading condition and the data are judged to be robust because of the consistency of the results within the same tub when applying multiple loading sequences, and between tubs when applying a similar test protocol. The experimental data from the test program will be used to update ExxonMobil suction pile design practice, and to validate in-house and contractor numerical models.