Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

Hossain, M. S., Hu, Y., Randolph, M. F. & White, D. J. (2005). Ge ´otechnique 5

Hossain, M. S., Hu, Y., Randolph, M. F. & White, D. J. (2005). Ge ´otechnique 55, No. 9, 679–690 679 Limiting cavity depth for spudcan foundations penetrating clay M. S. HOSSAIN*, Y. HU†, M. F. RANDOLPH* and D. J. WHITE‡ Centrifuge model tests and ﬁnite element (FE) analysis have been conducted to study the penetration of spudcan foundations in uniform clay with nominally constant strength with depth. In particular, the transition between shallow penetration, with soil heaving to the ground sur- face, and deep penetration, with a localised ﬂow-round mechanism, has been investigated. This transition governs the onset of back-ﬂow and hence the depth of soil lying on the installed spudcan, which in turn inﬂuences the bearing capacity and also the potential for suction to develop and hence the uplift capacity and moment resis- tance of the foundation. The maximum cavity depth above the spudcan prior to any back-ﬂow is therefore a critical issue for spudcan assessment in clay. In the centrifuge model tests, a half-spudcan model penetrating against a transparent window has been used to visualise the soil ﬂow mechanisms around the spudcan during penetration. The formation of a cavity above the spudcan is revealed by both centrifuge modelling and FE analysis. It is found that there are three distinct penetration mechanisms during spudcan installation: during initial penetration, an open cavity is formed with vertical walls; with further penetration, soil ﬂows partially around the spudcan into the cavity; during deep penetration, the spudcan is fully embedded and the soil ﬂow mechanism is entirely localised. Over the wide range of normalised soil strengths explored, the soil back-ﬂow in the second stage was shown to be due to a ﬂow failure that was triggered by the spudcan penetration and not by wall failure, that is, the collapse of the vertical sides of the soil cavity. This observation is supported by FE analysis. The cavity depth due to ﬂow failure is much shallower than the criterion for wall failure that is incorporated in current design guidelines. Instead, a new design chart and expression is suggested with the normalised cavity depth expressed as a function of the soil shear strength, normalised by the effective unit weight of the soil and the spudcan diameter. KEYWORDS: clays; footings/foundations; model tests; numer- ical modelling; offshore engineering; plasticity Des essais de mode `le centrifuge et des analyses d’e ´le ´ments ﬁnis (FE) ont e ´te ´ effectue ´s pour e ´tudier la pe ´ne ´tration de fondations ‘spudcan’ dans une argile uniforme avec une force nominale correspondant a ` la profondeur. En particu- lier, nous avons e ´tudie ´ la transition entre pe ´ne ´tration peu profonde - avec soule `vement de sol a ` la surface -, et pe ´ne ´tration profonde, avec un me ´canisme ﬂow-round loca- lise ´. Cette transition gouverne le de ´but du refoulement et de la `, la profondeur du sol sur les spudcan installe ´s, ce qui, a ` terme, inﬂuence la capacite ´ porteuse et le potentiel de succion a ` se de ´velopper et donc la capacite ´ de redresse- ment et la re ´sistance de moment de la fondation. La profondeur de cavite ´ maximum au-dessus des spudcan, avant tout refoulement, est donc un point critique dans l’e ´valuation des spudcan dans l’argile. Dans les essais de mode `le centrifuge, une moitie ´ de spudcan pe ´ne ´trant, de- vant une fene ˆtre transparente, a e ´te ´ utilise ´e pour visualiser le me ´canisme d’affaissement du sol autour des spudcan pendant la pe ´ne ´tration. La mode ´lisation et les analyses FE font apparaı ˆtre la formation d’une cavite ´ au-dessus des spudcan. Il y a trois me ´canismes de pe ´ne ´tration distincts pendant l’installation des spudcan : pendant la pe ´ne ´tra- tion initiale, une cavite ´ ouverte se forme avec des murs verticaux ; avec une pe ´ne ´tration plus pousse ´e, le sol s’af- faisse partiellement autour des spudcan dans la cavite ´ ; pendant une pe ´ne ´tration profonde, les spudcan sont com- ple `tement enfouis et le me ´canisme d’affaissement du sol est entie `rement localise ´. Dans la vaste gamme de re ´sis- tances normalise ´es que nous avons e ´tudie ´es, le refoulement du sol au second stade e ´tait du ˆ a ` un de ´faut d’e ´coulement qui a e ´te ´ de ´clenche ´ par la pe ´ne ´tration du spudcan et non par la de ´faillance des murs, c’est-a `-dire l’effondrement des co ˆte ´s verticaux de la cavite ´. Cette observation est conﬁrme ´e par les analyses FE. La profondeur de cavite ´ due a ` la de ´faillance de l’e ´coulement est bien moindre que le crite `re de de ´faillance des murs qui est incorpore ´ dans les directives de conception actuelles. Nous sugge ´rons un nouveau tableau de design et d’expression ou ` la profon- deur de cavite ´ normalise ´e serait exprime ´e comme fonction de la re ´sistance au cisaillement du sol, normalise ´e par le poids unitaire effectif du sol et le diame `tre des spudcan. INTRODUCTION Jack-up rigs are widely used in offshore oil and gas explora- tion and increasingly in temporary production and main- tenance work. A typical modern jack-up unit comprises a buoyant triangular platform supported by three independent k-lattice legs, each resting on a large inverted conical footing known as a spudcan (Y oung et al., 1984). These are gen- erally circular or polygonal in plan, with a shallow conical underside and a central spigot to give improved sliding resistance, as illustrated schematically in Fig. 1. Spudcan diameters in excess of 20 m have become common in post- 1980 designs. Before the commencement of the jack-up operation, the spudcans are preloaded through the k-lattice legs by pump- ing seawater into compartments within the hull. The preload causes the spudcans to penetrate into the seabed until the load on the spudcan is equilibrated by the resistance of the underlying soil. In soft soil, a spudcan may penetrate up to 2 or 3 diameters to reach equilibrium (Endley et al., 1981; Craig & Higham, 1985). The purpose of preloading is to penetrate the foundation sufﬁciently so that its resulting bearing capacity exceeds that required during extreme storm loading by an acceptable margin of safety. For a 50-year Manuscript received 10 August 2004; revised manuscript accepted 4 August 2005. Discussion on this paper closes on 2 May 2006, for further details see p. ii. * Centre for Offshore Foundation Systems (COFS), School of Civil and Resource Engineering, University of Western Australia. † Department of Civil Engineering, Curtin University of Technol- ogy, Western Australia. ‡ Schoﬁeld Centre, Department of Engineering, University of Cambridge, UK. Downloaded by [ Indian Institute of Technology - Bhubaneswar] on [11/06/20]. Copyright © ICE Publishing, all rights reserved. design storm, a common practice is to preload the founda- tion to twice the working vertical load. The ballast is then discharged and the hull is raised further to provide an adequate air-gap for subsequent operation. The penetration resistance of the spudcan during preload- ing, and its ability to withstand combined vertical, horizontal and moment loading during an extreme storm event, will be affected by the extent to which an open cavity forms above the spudcan during penetration, or whether back-ﬂow of soil occurs over the top of the spudcan. Current design guide- lines (SNAME, 1997) estimate the maximum depth of a cavity from solutions for the stability of an open hole, recommending conservative solutions by Meyerhof (1972), based on Rankine pressures, for soil with uniform shear strength, and upper-bound plasticity solutions of Britto & Kusakabe (1982, 1983) for normally consolidated, or lightly overconsolidated, soil where the shear strength increases markedly with depth. The guidelines also ignore any poten- tial beneﬁts of transient suctions beneath the spudcan during uplift or moment loading, owing to uncertainty in estimating the onset and degree of back-ﬂow. The purpose of this paper is, through a combination of experimental evidence from model tests and ﬁnite element analyses, to demonstrate that the current design approach is based on the wrong mechanism. Instead, the onset of back- ﬂow, and the consequent stable cavity depth above the spudcan, is shown to occur at a transition in the failure mechanism during penetration from one involving surface heave to one involving rotational ﬂow around the spudcan. Assessment of back-ﬂow In soft normally consolidated clay, ﬁeld inspection reports by divers (Kee & Ims, 1984) and centrifuge observations (Hossain et al., 2004b) have indicated a progressive inﬁlling of soil over the spudcan immediately after penetration of the widest part of the spudcan below the mudline. Eventually, the spudcan becomes fully embedded, with soil covering the entire top surface with no signiﬁcant cavity. This complete back-ﬂow negates the signiﬁcant bearing capacity contribu- tion from the overburden surcharge term, as this is balanced by the weight of soil resting on the spudcan. However, when a spudcan penetrates into overconsolidated clay with a sig- niﬁcant strength at the soil surface, uploads/Management/ geot-2005-55-9-679 1 .pdf