With the majority of estimated Arctic oil and gas reserves being held offshore, ice gouging will likely be a major consideration in the design of transport pipelines in these regions. The implications of the effects of ice gouging on buried pipelines are well understood. The ability to model this phenomenon using advanced numerical simulation tools has been proven in recent years, and is demonstrated in this paper. The uncertainty that revolves around these tools, due to limitations in the available physical dataset that can be utilized to validate the results, is discussed.
In this paper, a limited parametric study on the influence of ice keel attack angle and interface strength on the free-field subgouge displacement field, and subsequent effects on a buried pipeline is presented. The Coupled Eulerian Lagrangian finite element formulation available in ABAQUS/Explicit and the Arbitrary Lagrangian Eulerian formulation in LS-DYNA are used to conduct the numerical experiment. The results are shown and the observations are discussed in detail. Finally, an assessment in terms of the challenges of implementing the numerical tools in an engineering application is provided.
Energy demand has promoted renewed interest in the exploration and field development of offshore hydrocarbon basins in the arctic and ice covered waters of the northern hemisphere. In these harsh environments, pipelines offer a safe and cost effective mode to transport hydrocarbon resources to the marketplace. The presence of ice features and potential interaction with the seabed impose significant engineering challenges for design, construction and operation of subsea pipelines. Key technical issues relate to establishing pipeline mechanical performance criteria and trenching requirements for pipeline protection that meet target safety levels and satisfy logistical and economic constraints.
Ice gouge events involve ice keel/seabed interaction, soil failure mechanisms, soil clearing processes and subgouge soil deformations [1,2]. As the magnitude of ice keel/seabed reaction forces can be an order greater than other conventional load events; such as anchor dragging and pullover, then direct ice/pipeline contact is not a viable design option for most ice gouge scenarios [3]. In the 1970's, the use of extremal statistics was considered sufficient to estimate the design gouge depth and corresponding trench depth requirements. Later, in the 1990's, it was realized this approach to avoid direct ice/pipeline contact was incomplete. This conclusion was primarily based on physical modelling studies highlighting the importance of subgouge soil deformations and potential effects on the pipeline mechanical response [4,5]. Early continuum finite element (FE) modelling investigations on ice gouge events did not achieve the research objectives due to technology constraints [1,6]. Two-dimensional simulation models were not representative of the ice gouge clearing processes, and three-dimensional analysis encountered numerical instability due to excessive element distortion.