No abstract

A general failure mechanism, which is incorporated in an upper-bound analysis procedure and involves both continuous deformation and rigid-body rotation, has been developed. Analyses with this mechanism result in improved predictions for the design of slopes whose stability is affected by moderately high water pressures. In such cases, use of conventional analysis procedures may some-times lead to unsafe conditions. The mechanism was developed to explain a welldocumented slope failure at an operating lignite mine in north-east Texas. Analyses with the improved mechanism indicate that the slope was marginally stable, with a safety factor of 1·04, when it was being surveyed two days before it failed. Only a small increase in the water pressures, as indicated by a 0·5 m increase in the height of the water table near the slope face, was required for the safety factor to drop to 1·0. In contrast, both a conventional upper-bound analysis incorporating a rigid-body failure mechanism and a limiting-equilibrium analysis based on Spencer's method predict that the slope would be stable with a safety factor of 1·28 for the latter water pressure conditions. Comparisons of results with published solutions also indicate that the new mechanism can result in improved predictions of slope stability. The most significant improvement occurred in the analysis of the slide at Lodalen, where a stability analysis which used Bishop's method resulted in a safety factor of 1·05. In contrast, the new method produced a safety factor of 0·95. On a développé un méanisme généra1 de rupture lequel est incorporé dans une procédure d'analyse donnant les valeurs plus élevées et comprend la deformation continue aussi bien que la rotation de corps rigide. Des analyses effectuées avec ce mécanisme donnent des prédictions améliorées pour la construction de pentes dont la stabilité est influencée par des pressions d'eau de niveau modéré. Dans de tels cas I'emploi de procédures d'analyse conventionnelles peut quelquefois conduire à des conditions dangereuses. Ce méanisme a été développé pour expliquer une rupture de pente bien documentée dans une mine de lignite en exploitation au nerd-ouest du Texas. Des analyses effectées à l'aide du méanisme amélioré indiquent que la pente avait une stabilité marginale avec un facteur de sécuritié de 1·04 lors d'une inspection faite deux jours avant la rupture. Le facteur de sécurité s'est abaissié à 1·0 à la suite d'une petite augmentation des pressions d'eau, correspondant à un accroissement de 0·5 m dans le niveau de la nappe aquifère au voisinage de la surface de la pente. En contraste, une analyse conventionnelle donnant les valeurs plus élevées et incorporant un mécanisme de rupture de corps rigide et aussi une analyse d'équilibre limite basée sur la méthode de Spencer prédisent que la pente serait stable avec un facteur de sécuritié de 1·28 dans le cas des pressions d'eau indiquées. Des comparaisons des réultats avec des méthodes publiées indiquent aussi que le nouveau mécanisme peut fournir des prédictions améliorées pour la stabilité des pentes. L'amélioration la plus importante s'est révélée dans l'analyse d'un glissement à Lodalen, où une analyse de stabilité basée sur la méthode de Bishop a donné un facteur de sécurité de 1·05, tandis que la nouvelle méthode a produit un tel facteur de 0·95.

As the industry moves further offshore in the Arctic where open water seasons are small to non-existent, ice management will be a critical enabling technology to allow floating vessels to keep station (for drilling, tanker loading, well workovers or subsea equipment maintenance). This paper describes some of the key considerations for accomplishing the complex task of ice management and demonstrates a simulation-based means for testing the efficacy of various tactics against measured ice data. Examples are provided wherein published ice management fleet deployment scenarios are evaluated using measured ice drift and thickness time histories provided by the Canadian Department of Fisheries and Oceans for prospective drilling sites in the Canadian Beaufort. The results provide insight into the ice management fleet composition, fleet deployment strategies and frequency/duration of expected downtime due to ice conditions that would exceed the operating limits of the fleet. Introduction A major challenge of high-Arctic development lies in water depths exceeding about 100 m, where the traditional bottom-founded structures become impractical and stationary floating vessels are required for drilling and other key offshore operations. Unmanaged drifting sea ice can generate loads far beyond the capabilities of conventional station-keeping systems that use either dynamic positioning (DP) or anchored moorings. Ice management -- the process of protecting a stationary vessel in moving ice using icebreakers working upstream of the vessel to create a continuous channel of thoroughly broken-up floes (Figure 1) -- is required to reduce ice loads to manageable levels. Contrary to traditional icebreaker operations, where escort icebreakers exploit weak zones in the ice to create a channel for a transiting vessel, ice management for a stationary vessel must deal with whatever ice drifts across the fixed location. Amongst the challenges is that sea ice frequently drifts at speeds over 1 knot, which can make it difficult for a reasonable number of icebreakers to process the ice into sufficiently small floe sizes. Additionally, the ice drift heading varies constantly, with frequent changes of 180 degrees or more in a few hours, which challenges the fleet's ability to maintain the protected vessel within the managed ice channel created by icebreakers working updrift in the moving ice. Finally, some multi-year ice floes in the high-Arctic are too thick to be broken by even the largest conceivable icebreakers, so it is important to know the frequency with which these features may be encountered.

No abstract

This paper describes the determination of lateral soil displacement profiles surrounding a jackup rig spud can penetrating in soft clay. The displacement profiles were inferred from pile moment data obtained from centrifuge tests which modeled a spud can penetrating next to three instrumented piles. A beam column model that incorporates movable soil supports was employed to infer the soil displacements from the pile moment data. At distances as close as 0.5 spud can radii from the spud can edge, the soil displacements were found to be small (? 0.02 spud can radii). The inferred displacements were found to be consistent with physical measurements within the model. INTRODUCTION In soft clays, independent-leg jackup rig spud cans often penetrate to significant depths. This causes the displacement and remolding of large volumes of soil. Where jackups are employed in work overs of existing facilities or drilling new wells, this soil movement can place significant loads on in-place foundations. The problem of determining the magnitude and extent of soil movement surrounding the spud can is difficult. The soil displacements must also be converted into loads on the piles adjacent to the spud can. Existing analytical and numerical methods for evaluating penetration of objects in a semi-infinite medium do not appear well suited for determining the soil displacements in this case. A jackup rigspud can may penetrate to significant depths, but such depths may not be large in comparison to the spud can diameter. Therefore, surface effects may need to be taken into account. Also, unlike penetrating piles where laterally displaced soil moves primarily outward, a spud can may have soil flow around and above it as it penetrates. In order to address this problem, a series of centrifuge experiments have been initiated at the Geotechnical Centrifuge Center of the University of Cambridge. The objective of the tests is to simulate the penetration of a spud can adjacent to installed piles. Three piles instrumented with strain gauges along their length are used. The piles are spaced at increasing lateral distance away from the spud can edge. As the spud can penetrates, the moment distribution in each of the piles is recorded. The following sections contain a description of the test program and present some of the results. A method of analysis that uses a beam-column model with movable soil supports is also described. Using this model and the measured moment data, the associated optimum lateral soil displacement profiles were determined for various levels of spud can penetration and increasing distance from the spud can edge. The optimum soil displacement profiles were used to construct a generalized four-parameter model of displacement as a function of the spud can penetration and distance from the spud can. ANALYTICAL STUDIES In association with the data interpretation efforts reported in this paper, several analytical studies were also conducted. Limit analysis procedures were used to examine potential failure mechanisms of the spud can penetrating the soil. It was concluded that extensive remolding of the soil is localized laterally to within a spud can radius.